Dear all,
I wonder if anyone has come across this issue?
I'm calculating directly age/sex standardised hospital admission rates for
sub-national areas, using national totals as the standard population. Since
individuals can have multiple admissions, it seems that the usual methods
for calculating confidence intervals (I usually use -dstdize-) are not
appropriate, due to extra-Poisson variation.
I've found references to methods that account for this, neatly summarised
here: <http://www.doh.wa.gov/Data/Guidelines/ConfIntguide.htm#tth_sEc4.6>
I'm having trouble figuring out exactly how to implement these methods in
Stata, and just wondering if anyone has dealt with this issue and has a
sample do file that I might be able to adapt? Or an alternative method?
Thanks a lot
Ben
----------------------
Ben Wheeler
Research Fellow, University of Bristol
Department of Social Medicine
Department of Exercise, Nutrition & Health Sciences
& South West Public Health Observatory
e: [email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
