h_data_plot: Helper function to tidy survival fit data

In tern: Create Common TLGs Used in Clinical Trials

Helper function to tidy survival fit data

Description

Convert the survival fit data into a data frame designed for plotting within g_km.

This starts from the broom::tidy() result, and then:

• Post-processes the strata column into a factor.

• Extends each stratum by an additional first row with time 0 and probability 1 so that downstream plot lines start at those coordinates.

• Adds a censor column.

• Filters the rows before max_time.

Usage

h_data_plot(fit_km, armval = "All", max_time = NULL)

Arguments

fit_km	(survfit) result of survival::survfit().
armval	(string) used as strata name when treatment arm variable only has one level. Default is "All".
max_time	(numeric(1)) maximum value to show on x-axis. Only data values less than or up to this threshold value will be plotted (defaults to NULL).

Value

A tibble with columns time, n.risk, n.event, n.censor, estimate, std.error, conf.high, conf.low, strata, and censor.

Examples

library(dplyr)
library(survival)

# Test with multiple arms
tern_ex_adtte %>%
  filter(PARAMCD == "OS") %>%
  survfit(formula = Surv(AVAL, 1 - CNSR) ~ ARMCD, data = .) %>%
  h_data_plot()

# Test with single arm
tern_ex_adtte %>%
  filter(PARAMCD == "OS", ARMCD == "ARM B") %>%
  survfit(formula = Surv(AVAL, 1 - CNSR) ~ ARMCD, data = .) %>%
  h_data_plot(armval = "ARM B")

tern documentation built on June 22, 2024, 10:25 a.m.