1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/Rcode.r
\name{rsadd}
\alias{rsadd}
\title{Fit an Additive model for Relative Survival}
\usage{
rsadd(
formula = formula(data),
data = parent.frame(),
ratetable = relsurv::slopop,
int,
na.action,
method = "max.lik",
init,
bwin,
centered = FALSE,
cause,
control,
rmap,
...
)
}
\arguments{
\item{formula}{a formula object, with the response as a \code{Surv} object
on the left of a \code{~} operator, and, if desired, terms separated by the
\code{+} operator on the right. \code{Surv(start,stop,event)} outcomes
are also possible for time-dependent covariates and left-truncation for
\code{method='EM'}.
NOTE: The follow-up time must be in days.}
\item{data}{a data.frame in which to interpret the variables named in the
\code{formula}.}
\item{ratetable}{a table of event rates, organized as a \code{ratetable}
object, such as \code{slopop}.}
\item{int}{either a single value denoting the number of follow-up years or a
vector specifying the intervals (in years) in which the hazard is constant
(the times that are bigger than \code{max(int)} are censored. If missing,
only one interval (from time 0 to maximum observation time) is assumed. The
EM method does not need the intervals, only the maximum time can be
specified (all times are censored after this time point).}
\item{na.action}{a missing-data filter function, applied to the model.frame,
after any subset argument has been used. Default is
\code{options()$na.action}.}
\item{method}{\code{glm.bin} or \code{glm.poi} for a glm model, \code{EM}
for the EM algorithm and \code{max.lik} for the maximum likelihood model
(default).}
\item{init}{vector of initial values of the iteration. Default initial
value is zero for all variables.}
\item{bwin}{controls the bandwidth used for smoothing in the EM algorithm.
The follow-up time is divided into quartiles and \code{bwin} specifies a
factor by which the maximum between events time length on each interval is
multiplied. The default \code{bwin=-1} lets the function find an appropriate
value. If \code{bwin=0}, no smoothing is applied.}
\item{centered}{if \code{TRUE}, all the variables are centered before
fitting and the baseline excess hazard is calculated accordingly. Default is
\code{FALSE}.}
\item{cause}{A vector of the same length as the number of cases. \code{0}
for population deaths, \code{1} for disease specific deaths, \code{2}
(default) for unknown. Can only be used with the \code{EM} method.}
\item{control}{a list of parameters for controlling the fitting process.
See the documentation for \code{glm.control} for details.}
\item{rmap}{an optional list to be used if the variables are not organized
and named in the same way as in the \code{ratetable} object. See details
below.}
\item{...}{other arguments will be passed to \code{glm.control}.}
}
\value{
An object of class \code{rsadd}. In the case of
\code{method="glm.bin"} and \code{method="glm.poi"} the class also inherits
from \code{glm} which inherits from the class \code{lm}. Objects of this
class have methods for the functions \code{print} and \code{summary}. An
object of class \code{rsadd} is a list containing at least the following
components: \item{data}{the data as used in the model, along with the
variables defined in the rate table} \item{ratetable}{the ratetable used.}
\item{int}{the maximum time (in years) used. All the events at and after
this value are censored.} \item{method}{the fitting method that was used.}
\item{linear.predictors}{the vector of linear predictors, one per subject.}
}
\description{
The function fits an additive model to the data. The methods implemented are
the maximum likelihood method, the semiparametric method, a glm model with a
\code{binomial} error and a glm model with a \code{poisson} error.
}
\details{
NOTE: The follow-up time must be specified in days. The \code{ratetable}
being used may have different variable names and formats than the user's
data set, this is dealt with by the \code{rmap} argument. For example, if
age is in years in the data set but in days in the \code{ratetable} object,
age=age*365.241 should be used. The calendar year can be in any date format
(Date and POSIXt are allowed), the date formats in the
\code{ratetable} and in the data may differ.
The maximum likelihood method and both glm methods assume a fully parametric
model with a piecewise constant baseline excess hazard function. The
intervals on which the baseline is assumed constant should be passed via
argument \code{int}. The EM method is semiparametric, i.e. no assumptions
are made for the baseline hazard and therefore no intervals need to be
specified.
The methods using glm are methods for grouped data. The groups are formed
according to the covariate values. This should be taken into account when
fitting a model. The glm method returns life tables for groups specified by
the covariates in \code{groups}.
The EM method output includes the smoothed baseline excess hazard
\code{lambda0}, the cumulative baseline excess hazard \code{Lambda0} and
\code{times} at which they are estimated. The individual probabilites of
dying due to the excess risk are returned as \code{Nie}. The EM method
fitting procedure requires some local smoothing of the baseline excess
hazard. The default \code{bwin=-1} value lets the function find an
appropriate value for the smoothing band width. While this ensures an
unbiased estimate, the procedure time is much longer. As the value found by
the function is independent of the covariates in the model, the value can be
read from the output (\code{bwinfac}) and used for refitting different
models to the same data to save time.
}
\examples{
data(slopop)
data(rdata)
#fit an additive model
#note that the variable year is given in days since 01.01.1960 and that
#age must be multiplied by 365.241 in order to be expressed in days.
fit <- rsadd(Surv(time,cens)~sex+as.factor(agegr)+ratetable(age=age*365.241),
ratetable=slopop,data=rdata,int=5)
#check the goodness of fit
rs.br(fit)
#use the EM method and plot the smoothed baseline excess hazard
fit <- rsadd(Surv(time,cens)~sex+age,rmap=list(age=age*365.241),
ratetable=slopop,data=rdata,int=5,method="EM")
sm <- epa(fit)
plot(sm$times,sm$lambda,type="l")
}
\references{
Package. Pohar M., Stare J. (2006) "Relative survival analysis
in R." Computer Methods and Programs in Biomedicine, \bold{81}: 272--278
Relative survival: Pohar, M., Stare, J. (2007) "Making relative survival
analysis relatively easy." Computers in biology and medicine, \bold{37}:
1741--1749.
EM algorithm: Pohar Perme M., Henderson R., Stare, J. (2009) "An approach to
estimation in relative survival regression." Biostatistics, \bold{10}:
136--146.
}
\seealso{
\code{\link{rstrans}}, \code{\link{rsmul}}
}
\keyword{survival}
|
