Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: within stage of disease differences


From   "Austin Nichols" <[email protected]>
To   [email protected]
Subject   Re: st: within stage of disease differences
Date   Wed, 27 Jun 2007 16:08:58 -0400

Carlo--
I agree with Svend.  I would advise you to start by redefining
"failure" as movement from Stage 1 to Stage 2 instead of movement from
alive to dead.  Then repeat for each other transition--what you want
to do is measure the effect of a program on duration of a Stage, and
-streg- is the place to start.  You may find Stephen Jenkins's website
useful:
http://www.iser.essex.ac.uk/teaching/degree/stephenj/ec968/

I don't think you need enter the QALY fray just yet, not until you
want to "value" the different durations.

On 6/27/07, Svend Juul <[email protected]> wrote:
Carlo asked:

I beg Your pardon for the possible trivialism of what follows. I am intended
to compare two different health care programmes (A and B) which perform the
same in terms of overall average per patient survival (13 years each) but
give different results in terms of years patients (50,000 patients enrolled
in programme A and 50,000 in programme B) spend throughout different stages
(I^ II^ III^ and IV^) of the disease. Particularly, I would kindly ask You
about some hints to check for any statistical significance within each
stage. I tried tabi, chi2 exact with Stata 9, but I am not sure this the
right way to tackle this issue.

-----------------------------------------------------

This is far from trivial, and this type of question has lead health
economists, philosophers, epidemiologists, and biostatisticians to go war
against each other.

It looks like a QALY (Quality Adjusted Life Years) problem. To my knowledge
(and -findit-'s knowledge) no official or unoffical procedures deal with that.

Now, you ask about checking for statistical significance within each stage.
This requires data at the individual level with number of terminating events (stage shift, death) and time at risk. Look at survival analysis
(-st...- commands; the [ST] manual) and incidence rate ratios
(-ir-; also in [ST] under -epitab-).

Hope this helps (a bit)
Svend

*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index