add_median_survival: Calculate median survival time

In MBender1992/emR: Dictionary for EuMelaReg functions (analysis of clinical studies)

Calculate median survival time

Description

Fitting a survival fit for each subgroup defined by var and extracting median survival times. Additionally the total survival time for the whole sample is calculated. This function fits a survival curve for each subgroup defined by var and extracts the median survival times. Additionally the median survival time for the whole sample is calculated.

Usage

add_median_survival(
  data,
  time,
  status,
  var,
  round = 1,
  statistics = TRUE,
  weights = NULL,
  conf.type = "log-log"
)

Arguments

data	data.frame or data.table containing survival data.
time	The time interval from start of observation until date of event (e.g. disease progression or death) or censoring.
status	Variable specifying if event occured or data has been censored. Default behaviour inherited from the surv_fit function in the survival package, with 0 indicating censored data and 1 indicating event.
var	Variable tested for Influence on outcome.
round	rounds the results to the specified number of decimal places (default 1)
statistics	Logical value. If TRUE pvalue is printed. Default is TRUE. Statistical test is log-rank test.
weights	character variable specifying the name of the weights column. Weights (numerical vector) have to be added to the original dataframe in order to be applied correctly.
conf.type	Method to calculate confidence intervals. Log-log method is the default in SAS and was therefore set as the default here too.

Examples

# example survival calculation with the lung dataset
df <- survival::lung
add_median_survival(data = df, time = "time", status = "status", var = "ph.ecog")
# calculate median survival with inverse propensity score weighting for sex
df$sex <- factor(df$sex)
df$ph.ecog <- ifelse(is.na(df$ph.ecog), "Missing", df$ph.ecog)
df$ph.karno <- ifelse(is.na(df$ph.karno), "Missing", df$ph.karno)
df$weights.ate <- ate_weights(data = df, vars = c("age", "ph.ecog", "ph.karno"), prop.var = "sex")
# the internal logrank test (logrank_IPSW_RISCA) needs a status variable of 0 and 1
df$status <- ifelse(df$status == 2, 1, 0)
add_median_survival(data = df, time = "time", status = "status",
var = "sex", weights = "weights.ate")

MBender1992/emR documentation built on Feb. 18, 2025, 9:21 a.m.