md.survnp {missDeaths}

Nonparametric analysis of disease recurrence with missing information of possible deaths

Description

Estimates the Net and Event free survial using a is non-parametric approach that aims to correct all individuals using the unconditional survival time distribution obtained from the population mortality table.

The idea comes from realizing that the number of observed events in the data equals the number which would be observed in case of a complete data set, but the number of patients at risk does not. Hence, this method adjusts the observed number at risk to mimic the one we would get if the data was complete.

Usage

md.survnp(time, status, maxtime, D, ratetable, conf.int = 0.95)

Arguments

time

the time to event (number of days)

status

the status indicator, 0=right censored, 1=event at time

maxtime

maximum potential observation time (number of days).

where status=0 equals time.

where status=1 equals potential time of right censoring if no event would be observed.

D

demographic information compatible with ratetable, see md.D.

ratetable

a population mortality table, default is slopop

conf.int

desired coverage of the estimated confidence interval

Value

A list with components giving the estimates of net and event free survival.

time

times where the estimates are calculated (number of days)

Y.net

adjusted number of patients at risk at each time in a hypothetical world where patients don't die

Y.efs

adjusted number of patients at risk at each time

surv.net

the estimated Net survival

std.err.net

the estimated standard error of Net survival estimates

surv.efs

the estimated Event free survival

std.err.efs

the estimated standard error of Event free survival estimates

References

Stupnik T., Pohar Perme M. (2015) "Analysing disease recurrence with missing at risk information." Statistics in Medicine 35. p1130-43. https://onlinelibrary.wiley.com/doi/abs/10.1002/sim.6766

Examples

## Not run: 
library(missDeaths)
library(cmprsk)
data(slopop)

data(observed)
D = md.D(age=observed$age*365.2425, sex=observed$sex, year=(observed$year - 1970)*365.2425)
np = md.survnp(observed$time*365.2425, observed$status, observed$maxtime*365.2425, D, slopop)

#calculate net survival at 3 and 9 years
w = list(list(time=np$time, est=np$surv.net, var=(np$std.err.net)^2))
timepoints(w, times=c(3,9)*365.2425)

#plot the net and event free survival curves
plot(np$time, np$surv.net)
plot(np$time, np$surv.efs)

## End(Not run)