On 06/09/2013 11:14 PM, Calum Polwart wrote:
...
What we are trying to do is determine the most appropriate number to
make the capsules. (Our dosing is more complex but lets stick to
something simple. I can safely assure you that vritually no-one actually
needs 250 or 500mg as a dose of amoxicillin... ...thats just a dose to
get them into a therapeutic window, and I'm 99% certain 250 and 500 are
used coz they are round numbers. if 337.5 more reliably got everyone in
the window without kicking anyone out the window that'd be a better dose
to use! So... what I'm looking to do is model the 'theoretical dose
required' (which we know) and the dose delivered using several starting
points to get the 'best fit'. We know they need to be within 7% of each
other, but if one starting point can get 85% of doses within 5% we think
that might be better than one that only gets 50% within 5%.
Okay, I think I see what you are attempting now. You are stuck with
fairly large dosage increments (say powers of two) and you want to have
a "base" value that will be appropriate for the greatest number of
patients. So, your range of doses can be generated with:
d * 2 ^ (0:m)
where d is some constant and m+1 is the number of doses you want to
generate. For your amoxcillin, d=250 and m=1, so you get 250 and 500mg.
Given this relationship (or any other one you can define), you want to
set your base dose so that it is close to the mode of the patient
distribution. This means that the greatest number of patients will be
suitably dosed with your base dose. I would probably try to solve this
by brute force, setting the base dose at the mode and then moving it up
and down until the dose was appropriate for the largest number of patients.
However, there are a lot of people on this list who would be more
familiar with this sort of problem, and there may be a more elegant
solution.
Jim
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.