Lexis.diagram: Plot a Lexis diagram

Description

Draws a Lexis diagram, optionally with life lines from a cohort, and with lifelines of a cohort if supplied. Intended for presentation purposes.

Usage

Lexis.diagram( age = c( 0, 60), 
              alab = "Age",
              date = c( 1940, 2000 ),
              dlab = "Calendar time",
               int = 5,
           lab.int = 2*int, 
          col.life = "black",
          lwd.life = 2,
          age.grid = TRUE,
         date.grid = TRUE,
          coh.grid = FALSE,
          col.grid = gray(0.7),
          lwd.grid = 1,     
               las = 1,
        entry.date = NA,
         entry.age = NA,
         exit.date = NA,
          exit.age = NA,
         risk.time = NA,
        birth.date = NA,
              fail = NA,
          cex.fail = 1.1, 
          pch.fail = c(NA,16),
          col.fail = rep( col.life, 2 ),
              data = NULL, ... )

Arguments

age

Numerical vector of length 2, giving the age-range for the diagram

alab

Label on the age-axis.

date

Numerical vector of length 2, giving the calendar time-range for the diagram

dlab

label on the calendar time axis.

int

The interval between grid lines in the diagram. If a vector of length two is given, the first value will be used for spacing of age-grid and the second for spacing of the date grid.

lab.int

The interval between labelling of the grids.

col.life

Colour of the life lines.

lwd.life

Width of the life lines.

age.grid

Should grid lines be drawn for age?

date.grid

Should grid lines be drawn for date?

coh.grid

Should grid lines be drawn for birth cohorts (diagonals)?

col.grid

Colour of the grid lines.

lwd.grid

Width of the grid lines.

las

How are the axis labels plotted?

entry.date, entry.age, exit.date, exit.age, risk.time, birth.date

Numerical vectors defining lifelines to be plotted in the diagram. At least three must be given to produce lines. Not all subsets of three will suffice, the given subset has to define life lines. If insufficient data is given, no life lines are produced.

fail

Logical of event status at exit for the persons whose life lines are plotted.

pch.fail

Symbols at the end of the life lines for censorings (fail==0) and failures (fail != 0).

cex.fail

Expansion of the status marks at the end of life lines.

col.fail

Character vector of length 2 giving the colour of the failure marks for censorings and failures respectively.

data

Dataframe in which to interpret the arguments.

...

Arguments to be passed on to the initial call to plot.

Value

If sufficient information on lifelines is given, a data frame with one row per person and columns with entry ages and dates, birth date, risk time and status filled in.

Side effect: a plot of a Lexis diagram is produced with the life lines in it is produced. This will be the main reason for using the function. If the primary aim is to illustrate follow-up of a cohort, then it is better to represent the follow-up in a Lexis object, and use the generic plot.Lexis function.

Details

The default unit for supplied variables are (calendar) years. If any of the variables entry.date, exit.date or birth.date are of class "Date" or if any of the variables entry.age, exit.age or risk.time are of class "difftime", they will be converted to calendar years, and plotted correctly in the diagram. The returned dataframe will then have colums of classes "Date" and "difftime".

Examples

Run this code

# NOT RUN {
Lexis.diagram( entry.age = c(3,30,45),
               risk.time = c(25,5,14),
              birth.date = c(1970,1931,1925.7),
                    fail = c(TRUE,TRUE,FALSE) )
LL <- Lexis.diagram( entry.age = sample( 0:50, 17, replace=TRUE ),
                     risk.time = sample( 5:40, 17, r=TRUE),
                    birth.date = sample( 1910:1980, 17, r=TRUE ),
          		  fail = sample( 0:1, 17, r=TRUE ), 
	              cex.fail = 1.1,
	              lwd.life = 2 )
# Identify the persons' entry and exits
text( LL$exit.date, LL$exit.age, paste(1:nrow(LL)), col="red", font=2, adj=c(0,1) )
text( LL$entry.date, LL$entry.age, paste(1:nrow(LL)), col="blue", font=2, adj=c(1,0) )
data( nickel )
attach( nickel )
LL <- Lexis.diagram( age=c(10,100), date=c(1900,1990), 
	             entry.age=age1st, exit.age=ageout, birth.date=dob, 
		     fail=(icd %in% c(162,163)), lwd.life=1,
		     cex.fail=0.8, col.fail=c("green","red") )
abline( v=1934, col="blue" )
nickel[1:10,]
LL[1:10,]
# }