Add p-value to a survival table

Calculates and adds a p-value comparing the survival curves of each stratified model in a table created with tbl_survfit(). Called as add_p() on a tbl_survfit object.

Value

A tbl_survfit table with a p.value column.

Arguments

• x (tbl_survfit)
  A table created with tbl_survfit().

• test (string or formula-list)
  The test: one of "logrank", "survdiff", "petopeto_gehanwilcoxon", "coxph_lrt", "coxph_wald" and "coxph_score". A string applies to every model; a formula list sets the test per variable, e.g. list(trt ~ "logrank", grade ~ "survdiff"). Default is "logrank". See tests for how each is computed.

• test.args (named list or formula-list)
  Additional arguments passed to the test, e.g. list(rho = 0.5) for a test of the G-rho family with test = "survdiff". A named list applies to every model; a formula list sets the arguments per variable.

• pvalue_fun (function)
  Function that rounds and formats the p-values. Default is label_style_pvalue(digits = 1).

• include (selector)
  The stratified models to compute a p-value for. Default is everything().

• ... Not used.

Note

The p-values are computed by survival::survdiff() or survival::coxph() from the formula and data of the original survfit() call, so both must be evaluable; see the formula specification section of tbl_survfit() and tbl_survfit_errors.

See also

Other survival tables: add_n_survfit, add_nevent_survfit, tbl_survfit(), tbl_survfit_errors

Examples

library(survival)

tbl <- list(
  survfit(Surv(ttdeath, death) ~ grade, trial),
  survfit(Surv(ttdeath, death) ~ trt, trial)
) |>
  tbl_survfit(times = c(12, 24))

# Example 1 ----------------------------------
tbl |>
  add_p()

# Example 2 ----------------------------------
# a test of the G-rho family, with `rho` passed to `survdiff()`
tbl |>
  add_p(test = "survdiff", test.args = list(rho = 0.5))