This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Survival analysis in R keeps censoring figure with no censored data.

I am following THIS guide on how to create a survival analysis in R.

The instructions say, "Each row is a subject. An event (death) is indicated if the variable outcome is 1, in which case time indicates the time of death. An outcome of 0 refers to no event (censored data), in which case time is the time of censoring.".

However, I have datasets which have no censored rows as all events were deaths but when I produce my figure my samples are shaded as if censored.

Example of 2 samples which contain all deaths and no censored events.

This is the code I have been running:

#TWO GROUP KAPLAN-MEIER CURVE
#read in file
ATCC <- read.csv("ATCCsample.csv")
head(ATCC )

#Surv beeded to convert the data to a format used by package
ATCCSurv <- Surv(time = ATCC$OverallSurvivalDays, event = ATCC$status)
#Survfit used to calculate survival curve
ATCCKM <- survfit(ATCCSurv ~ Subtype, data = ATCC , type="kaplan-meier") 

#plot curve
ggsurvplot(ATCCKM, conf.int = TRUE, pval = FALSE, risk.table = FALSE,
  legend.labs=c("ATCC  1", "ATCC  2"), 
  legend.title = "Subtype",
  censor.shape = "|", censor.size = 4,
  palette=c("firebrick", "goldenrod1"), 
  xlab = "Time since inoculation (days)",
  ylab = "Proportion surviving")

I have tried playing around with the "censor.size" but it doesn't look any different whether I change it to 0 or 20. It produces the same figure.

Is there something wrong with the code I am using? In THIS data set, each row only has a 1, and those still alive by day 7 I left blank (did not input 1 or 0).

r survival-analysis censor

0 answers

No answers yet.

Log in to answer this question.