Functions to facilitate analysis of multistate models.

Description

stack.Lexis produces a stacked object suited for analysis of several transition intensities simultaneously.

Usage

## S3 method for class 'Lexis'
stack(x, ...)
tmat( x, ... )
## S3 method for class 'Lexis'
tmat(x, Y=FALSE, mode = "numeric", ...)

Arguments

Value

tmat.Lexis returns a square transition matrix, classified by the levels of lex.Cst and lex.Xst, for every transition occurring the entry is the number of transitions occurring and NA in all oter entries. If Y=TRUE, the diagonal will contain the risk time in each of the states.

stack.Lexis returns a dataframe to be used for analysis of multistate data when all transitions are modelled together, for example if some parameters are required to be the same for different transitions. The dataframe has class stacked.Lexis, and inherits the attributes time.scales and breaks from the Lexis object, and so function timeBand applies to a stacked.Lexis object too.

The dataframe has same variables as the original Lexis object, but with each record duplicated as many times as there are possible exits from the current state, lex.Cst. Two variables are added: lex.Fail, an indicator of wheter an event for the transition named in the factor lex.Tr has occurred or not. lex.Tr is a factor with levels made up of combinations of the levels of lex.Cst and lex.Xst that do occur together in x, joined by a "->".

Author(s)

Bendix Carstensen, bxc@steno.dk, http://BendixCarstensen.com

See Also

splitLexis cutLexis Lexis

Examples

data(DMlate)
str(DMlate)
dml <- Lexis( entry=list(Per=dodm, Age=dodm-dobth, DMdur=0 ),
               exit=list(Per=dox),
        exit.status=factor(!is.na(dodth),labels=c("DM","Dead")),
               data=DMlate )
dmi <- cutLexis( dml, cut=dml$doins, new.state="Ins", pre="DM" )
summary( dmi )
ls.dmi <- stack( dmi )
str( ls.dmi )
# Check that all the transitions and person-years got across.
with( ls.dmi, rbind( table(lex.Fail,lex.Tr),
                     tapply(lex.dur,lex.Tr,sum) ) )

Results

R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(Epi)

Attaching package: 'Epi'

The following object is masked from 'package:base':

    merge.data.frame

> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/Epi/stack.Lexis.Rd_%03d_medium.png", width=480, height=480)
> ### Name: stack.Lexis
> ### Title: Functions to facilitate analysis of multistate models.
> ### Aliases: stack.Lexis tmat tmat.Lexis
> ### Keywords: survival
> 
> ### ** Examples
> 
> data(DMlate)
> str(DMlate)
'data.frame':	10000 obs. of  7 variables:
 $ sex  : Factor w/ 2 levels "M","F": 2 1 2 2 1 2 1 1 2 1 ...
 $ dobth: num  1940 1939 1918 1965 1933 ...
 $ dodm : num  1999 2003 2005 2009 2009 ...
 $ dodth: num  NA NA NA NA NA ...
 $ dooad: num  NA 2007 NA NA NA ...
 $ doins: num  NA NA NA NA NA NA NA NA NA NA ...
 $ dox  : num  2010 2010 2010 2010 2010 ...
> dml <- Lexis( entry=list(Per=dodm, Age=dodm-dobth, DMdur=0 ),
+                exit=list(Per=dox),
+         exit.status=factor(!is.na(dodth),labels=c("DM","Dead")),
+                data=DMlate )
NOTE: entry.status has been set to "DM" for all.
Warning message:
In Lexis(entry = list(Per = dodm, Age = dodm - dobth, DMdur = 0),  :
  Dropping 4 rows with duration of follow up < tol

> dmi <- cutLexis( dml, cut=dml$doins, new.state="Ins", pre="DM" )
> summary( dmi )
     
Transitions:
     To
From    DM  Ins Dead  Records:  Events: Risk time:  Persons:
  DM  6157 1694 2048      9899     3742   45885.49      9899
  Ins    0 1340  451      1791      451    8387.77      1791
  Sum 6157 3034 2499     11690     4193   54273.27      9996
> ls.dmi <- stack( dmi )
> str( ls.dmi )
Classes 'stacked.Lexis' and 'data.frame':	21589 obs. of  16 variables:
 $ Per     : num  1999 2003 2005 2009 2009 ...
 $ Age     : num  58.7 64.1 86.3 44 75.8 ...
 $ DMdur   : num  0 0 0 0 0 0 0 0 0 0 ...
 $ lex.dur : num  11.08 6.689 5.446 0.736 1.344 ...
 $ lex.Cst : Factor w/ 3 levels "DM","Ins","Dead": 1 1 1 1 1 1 1 1 1 1 ...
 $ lex.Xst : Factor w/ 3 levels "DM","Ins","Dead": 1 1 1 1 1 3 1 1 3 1 ...
 $ lex.Tr  : Factor w/ 3 levels "DM->Ins","DM->Dead",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ lex.Fail: logi  FALSE FALSE FALSE FALSE FALSE FALSE ...
 $ lex.id  : int  1 2 3 4 5 6 7 8 9 10 ...
 $ sex     : Factor w/ 2 levels "M","F": 2 1 2 2 1 2 1 1 2 1 ...
 $ dobth   : num  1940 1939 1918 1965 1933 ...
 $ dodm    : num  1999 2003 2005 2009 2009 ...
 $ dodth   : num  NA NA NA NA NA ...
 $ dooad   : num  NA 2007 NA NA NA ...
 $ doins   : num  NA NA NA NA NA NA NA NA NA NA ...
 $ dox     : num  2010 2010 2010 2010 2010 ...
 - attr(*, "breaks")=List of 3
  ..$ Per  : NULL
  ..$ Age  : NULL
  ..$ DMdur: NULL
 - attr(*, "time.scales")= chr  "Per" "Age" "DMdur"
> # Check that all the transitions and person-years got across.
> with( ls.dmi, rbind( table(lex.Fail,lex.Tr),
+                      tapply(lex.dur,lex.Tr,sum) ) )
       DM->Ins DM->Dead Ins->Dead
FALSE  8205.00  7851.00  1340.000
TRUE   1694.00  2048.00   451.000
      45885.49 45885.49  8387.773
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>