Debian Codegraph @

man/msfit.Rd

raw
1 % Generated by roxygen2: do not edit by hand
2 % Please edit documentation in R/msfit.R
3 \name{msfit}
4 \alias{msfit}
5 \title{Compute subject-specific transition hazards with (co-)variances}
6 \usage{
7 msfit(
8 object,
9 newdata,
10 variance = TRUE,
11 vartype = c("aalen", "greenwood"),
12 trans
13 )
14 }
15 \arguments{
16 \item{object}{A \code{\link[survival:coxph]{coxph}} object describing the
17 fit of the multi-state model}
18
19 \item{newdata}{A data frame with the same variable names as those that
20 appear in the \code{coxph} formula. Its use is somewhat different from
21 \code{\link[survival:survfit]{survfit}}. See Details. The argument
22 \code{newdata} may be omitted only if the right hand side of the formula in
23 the \code{coxph} object is \code{~strata(trans)}}
24
25 \item{variance}{A logical value indicating whether the (co-)variances of the
26 subject-specific transition hazards should be computed. Default is
27 \code{TRUE}}
28
29 \item{vartype}{A character string specifying the type of variances to be
30 computed (so only needed if \code{variance}=\code{TRUE}). Possible values
31 are \code{"aalen"} or \code{"greenwood"}}
32
33 \item{trans}{Transition matrix describing the states and transitions in the
34 multi-state model. See \code{trans} in \code{\link{msprep}} for more
35 detailed information}
36 }
37 \value{
38 An object of class \code{"msfit"}, which is a list containing
39 \item{Haz }{A data frame with \code{time}, \code{Haz}, \code{trans},
40 containing the estimated subject-specific hazards for each of the
41 transitions in the multi-state model} \item{varHaz }{A data frame with
42 \code{time}, \code{Haz}, \code{trans1}, \code{trans2} containing the
43 variances (\code{trans1}=\code{trans2}) and covariances
44 (\code{trans1}<\code{trans2}) of the estimated hazards. This element is only
45 returned when \code{variance}=\code{TRUE}} \item{trans}{The transition
46 matrix used}
47 }
48 \description{
49 This function computes subject-specific or overall cumulative transition
50 hazards for each of the possible transitions in the multi-state model. If
51 requested, also the variances and covariances of the estimated cumulative
52 transition hazards are calculated.
53 }
54 \details{
55 The data frame needs to have one row for each transition in the multi-state
56 model. An additional column \code{strata} (numeric) is needed to describe
57 for each transition to which stratum it belongs. The name has to be
58 \code{strata}, even if in the original \code{coxph} call another variable
59 was used. For details refer to de Wreede, Fiocco & Putter (2010). So far,
60 the results have been checked only for the \code{"breslow"} method of
61 dealing with ties in \code{\link[survival:coxph]{coxph}}, so this is
62 recommended.
63 }
64 \examples{
65
66 # transition matrix for illness-death model
67 tmat <- trans.illdeath()
68 # data in wide format, for transition 1 this is dataset E1 of
69 # Therneau & Grambsch (2000)
70 tg <- data.frame(illt=c(1,1,6,6,8,9),ills=c(1,0,1,1,0,1),
71 dt=c(5,1,9,7,8,12),ds=c(1,1,1,1,1,1),
72 x1=c(1,1,1,0,0,0),x2=c(6:1))
73 # data in long format using msprep
74 tglong <- msprep(time=c(NA,"illt","dt"),status=c(NA,"ills","ds"),
75 data=tg,keep=c("x1","x2"),trans=tmat)
76 # events
77 events(tglong)
78 table(tglong$status,tglong$to,tglong$from)
79 # expanded covariates
80 tglong <- expand.covs(tglong,c("x1","x2"))
81 # Cox model with different covariate
82 cx <- coxph(Surv(Tstart,Tstop,status)~x1.1+x2.2+strata(trans),
83 data=tglong,method="breslow")
84 summary(cx)
85 # new data, to check whether results are the same for transition 1 as
86 # those in appendix E.1 of Therneau & Grambsch (2000)
87 newdata <- data.frame(trans=1:3,x1.1=c(0,0,0),x2.2=c(0,1,0),strata=1:3)
88 msfit(cx,newdata,trans=tmat)
89
90 }
91 \references{
92 Putter H, Fiocco M, Geskus RB (2007). Tutorial in biostatistics:
93 Competing risks and multi-state models. \emph{Statistics in Medicine}
94 \bold{26}, 2389--2430.
95
96 Therneau TM, Grambsch PM (2000). \emph{Modeling Survival Data: Extending the
97 Cox Model}. Springer, New York.
98
99 de Wreede LC, Fiocco M, and Putter H (2010). The mstate package for
100 estimation and prediction in non- and semi-parametric multi-state and
101 competing risks models. \emph{Computer Methods and Programs in Biomedicine}
102 \bold{99}, 261--274.
103
104 de Wreede LC, Fiocco M, and Putter H (2011). mstate: An R Package for the
105 Analysis of Competing Risks and Multi-State Models. \emph{Journal of
106 Statistical Software}, Volume 38, Issue 7.
107 }
108 \seealso{
109 \code{\link{plot.msfit}}
110 }
111 \author{
112 Hein Putter \email{[email protected]}
113 }
114 \keyword{survival}