Dear Hua,
One approach: use the function event.history in the Hmisc library.
A citation for a 2001 Statistics in Medicine paper (Dubin, Muller, Wang)
describing the method for plotting survival data with time-dependent
covariates is contained in the help file.
Good luck, Joel.
Joel Dubin
Departments of Statistics & Actuarial Science and Health Studies &
Gerontology
University of Waterloo
jdubin at uwaterloo.ca
_______________________
Message: 44
Date: Thu, 11 May 2006 12:35:52 -0500
From: Hua Ding <hua.ding at vanderbilt.edu>
Subject: [R] time-dependent covariate survival curves
To: r-help at stat.math.ethz.ch
Message-ID: <446375F8.3060102 at vanderbilt.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Dear r-users,
Does anyone know how to draw time-dependent survival curves?
Example:
Event outcome: CHD
Time-dependent covariate: NSAID use, which changes over time for each
subject
I'm interested in survival curves stratified by NSAID use.
I'd like to implement Simon & Makuch (1984) method. Is there a R
package/function to draw this graph?
------------------------------------------
Two related papers:
Simon R, Makuch RW. A non-parametric graphical representation of the
relationship between survival and the occurrence of an event:
application to responder versus non-responder bias. Statistics in
Medicine 1984; 3: 35-44.
Schultz LR, Peterson EL, Breslau N. Graphing survival curve estimates
for time-dependent covariates. Int'l J of Methods in Psychiatric
Research 2002; 11: 68-74.
--------------------------------------------------
-- Rebecca Ding Biostatistician II School of Medicine Department of
Biostatistics Vanderbilt University (W)615-343-3878