Kaplan-Meier plot code (ggsurvfit)

Usage

tfl_fig_km(
  adam = NULL,
  param = "OS",
  group = "TRT01P",
  style = c("risk_table", "simple", "ci", "single_arm"),
  legend = NULL,
  pop = "FASFL",
  data = "ADTTE",
  time = "AVAL",
  censor = "CNSR",
  time_unit = "months",
  title = NULL,
  file = NULL,
  plot_id = "km",
  ard = NULL,
  ...
)

Arguments

adam: Optional ADaM data (tfl_read_adam()); values found in the data (groups, colours) are written literally into the code.
param: PARAMCD of the time-to-event parameter.
group: Grouping variable (NULL for one curve).
style: One of tfl_fig_types() for km.
legend: none, right, bottom, inside, inside_bl, panel, panel_right, panel_inside (panel* = legend drawn from an item table).
pop: Population flag (== "Y"); NULL for none.
data: Dataset name.
time, censor: Time and censor (1 = censored) variables.
time_unit: Unit of time in the data is days; the axis is shown in days, weeks, months or years.
title: Figure title.
file: Write the script to this file.
plot_id: Used in the output file name.
ard: R code (a string) that gives the ARD of the KM table – cardx::ard_survival_survfit(times = ) – e.g. 'readRDS("output/ard/ard.rds") |> subset(output_id == "T-14-2-2")'. The number at risk is then read from it, so the figure and the table agree, and checked against the curve's own count. The risk table's times are the ARD's; the ARD's time unit must be the axis's.
...: Engine options, e.g. x_max = 24, x_by = 3, palette = "grey", theme = "classic", width = 7, height = 5.

Value

A tfl_code object (character; printed as the script).