Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

re:st: Cox "logrank" test after propensity score weighting


From   "Ariel Linden. DrPH" <ariel.linden@gmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   re:st: Cox "logrank" test after propensity score weighting
Date   Mon, 27 Feb 2012 09:06:50 -1000

Hi Adam,

The brains behind the IPTW framework in epidemiology research is Jamie
Robins (although he does a lot of work together with his colleague at
Harvard - Miguel Hernan). His standard approach to survival analysis with
weighting is pooled logistic regression. The unweighted  odds ratios are
equivalent to the hazard ratios that would be obtained from the equivalent
Cox model. 

Below are some references. The first one deals specifically with survival
analysis, whereas the other papers are more general and treat survival as a
special case of the approach.

I hope this helps

Ariel


Hernan, M. A., B. Brumback, and J. M. Robins. 2000. Marginal structural
models to
estimate the causal effect of zidovudine on the survival of HIV-positive
men. Epidemiology
11(5): 561-570.

Robins, J. M. 1999. Marginal structural models versus structural nested
models as tools
for causal inference. In Statistical Models in Epidemiology: The Environment
and
Clinical Trials, ed. M. E. Halloran and D. Berry, 95-134. New York:
Springer.

Robins, J. M., M. A. Hernan, and B. Brumback. 2000. Marginal structural
models and
causal inference in epidemiology. Epidemiology 11(5): 550-560.

Date: Sun, 26 Feb 2012 17:46:39 -0500
From: Adam Olszewski <adam.olszewski@gmail.com>
Subject: st: Cox "logrank" test after propensity score weighting

Hello,
I am comparing two treatment groups' survival after IPTW weighting on
a propensity score. The -sts test treatmentvar- command uses a
Cox-regression based comparison in this situation. I was wondering if
anyone has a citation that can be used to support this methodology
(other than the general statement that logrank test is not appropriate
with pweights) as used successfully in literature.
Wouldn't it be more appropriate to just run a multivariate Cox model
with other outcome-related variables?
Is this command still valid if the treatment variable does not meet
the PH assumption criteria?

Thanks for any insights,
Adam


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index