I'm not a spatial epidemiologist, but I'd like to learn more about how
these particular creatures resolve their analytical issues. What I'd like
to do is learn more about how spatial epi folks handle the situation of
needing to smooth disease risk across a set of units (say, counties within
a state) given no information on the number of infected relative to the
number of uninfected - in those cases when all they have is the proportion
(say, from a historical atlas or some other summarizing document that has
hidden the underlying sampling process). So, for instance, if you had
information that suggested various counties had some proportion of their
population with a certain disease, then how might you relate that to
covariates? You might think that this is a rather common situation, but I
am not aware of examples from the spatial epi literature (admittedly, my
perspective on it has been limited heretofore). Most examples start with
the premise that you have the number of infected relative to, say, the
number tested. This sort of approach has been modeled as a Poisson
process or a binomial process. If anyone can point me to spatial epi
examples where the underlying number of infected versus tested information
is absent, I would appreciate it. I just realized this isn't WinBUGS
specific, but it's the means by which I most often handle these sorts of
analyses so it seemed natural to inquire with folks of a similar nature.
Wayne Thogmartin
U.S. Geological Survey
La Crosse, WI
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|