Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Competing risks


From   Tero T Kivela <tekivela@cc.helsinki.fi>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Competing risks
Date   Tue, 23 Mar 2004 14:32:51 +0200 (EET)

On Tue, 22 Mar 2004, May Boggess wrote:

> On Mondy, Sarah Senesky wrote:
> > I would like to estimate a duration model with competing risks.  Is this
> > possible in Stata?  Could anyone direct me to information about how to run
> > models with competing risks?
> >
>
> There is an FAQ on the topic:
> http://www.stata.com/support/faqs/stat/stmfail.html
>
> Scroll down a little until you get to the section on competing risks.
> This section describes how to do a Cox Model with competing risks in
> Stata.

This is an interesting FAQ. However:

"We illustrate the use of Stata in the analysis of a competing-risk model,
with a subset of the Mayo Clinic's Ursodeoxycholic acid (UDCA) data
(Lindor et al. 1994). The data consists of 170 patients with primary
biliary cirrhosis randomly allocated to either the UDCA treatment group or
a group receiving a placebo. The times up to nine possible events were
recorded: death, liver transplant, voluntary withdraw, histologic
progression, development of varices, development of ascites, development
of encephalophathy, doubling of bilirubin, and worsening of symptoms. All
times were measured from the date of treatment allocation."

Most of these do not seem to be true competing risks - I understand that a
competing risk is an event that renders the subject "immune" to the event
of interest, e.g. a subject who dies can no longer receive a liver
transplant. For this purpose, cumulative incidence analysis and competing
risks proportional hazards regression of Gray have been developed.

I can not see that development of ascites and most other event mentioned
in the example are competing risks. They are multiple failure events of
interest, but nor competing risks as far as I can see.

Furthermore:

"An important characteristic of these failure events is that each can
occur only once per subject. Note that all subjects are at risk for all
events. Also, when a subject experiences one of the events, he remains at
risk for all other events."

Certainly, a subject who dies is not longer at risk of the other events.

I am puzzled. Primarily I believe that "competing risk" is a misnomer in
the FAQ title 3.1.2.

T. Kivela
Univ. of Helsinki
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index