R: st: Is there a similar command as roccomp for comparing two kaplan-meier curves?

["Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>]

Dear Ton,
what follows might be of some help:
--------------------------------------------------------------
use http://www.stata-press.com/data/r12/stan3
g A=2 if surgery==1 & transplant==1
replace A=1 if surgery==0 & transplant==1
replace A=0 if surgery==0 & transplant==0
sts graph, by(A)
sts test A, logrank
--------------------------------------------------------------

Take a look at - help sts test - for further details.

Kindest Regards,
Carlo



Dott. Carlo Lazzaro
Studio di Economia Sanitaria
Via Stefanardo da Vimercate, 19
20128 Milano
Tel/fax: 02/26000516
Portatile: 335/6786741
e-mail: carlo.lazzaro@tiscalinet.it
               carlo.lazzaro@tin.it

-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di
A.J.M.de_Craen@lumc.nl
Inviato: lunedì 8 ottobre 2012 10:00
A: statalist@hsphsun2.harvard.edu
Oggetto: RE: st: Is there a similar command as roccomp for comparing two
kaplan-meier curves?

Thank you Steve, I was indeed not completely clear. Let me retry.
For 600 individuals I have calculated two measures of kidney function based
on different blood parameters. First, I select those with a poor kidney
function based on one variable, and run a survival analysis. Then I select
those with a poor kidney function on the other variable, and run a survival
analysis again. Obviously, there are many individuals in both analyses, but
some only have a poor kidney function on one variable, and others only on
the other. When I look at the two Kaplan-Meier curves in one graph, I see a
clear difference between the two lines. How do I test whether they are
statistically significant from each other?
Regards,
Ton



-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Steve Samuels
Sent: maandag 8 oktober 2012 4:02
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Is there a similar command as roccomp for comparing two
kaplan-meier curves?



Your question is not clear. There cannot be two KM curves that both are
based on the same individuals, as the KM curve is unique, given the sample.
As asked in the FAQ Section 3.3, please show us the Stata commands that
produce the KM curves that you wish to compare. If you can, illustrate the
problem in one the data sets referenced in the
-help- for the command.


Steve


On Oct 7, 2012, at 3:32 PM, <A.J.M.de_Craen@lumc.nl>
<A.J.M.de_Craen@lumc.nl> wrote:

Is there a similar command as roccomp for comparing two kaplan-meier curves?
I want to compare the predictive value of two different variables in the
same dataset. Note that the all individuals are in both K-M curves.
Thanks,
Ton de Craen
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/