[Feature Request]: tbl_coxph

[reesnj]

Feature description

Suggestion for a new function tbl_coxph which computes the pairwise cox proportional hazards test and returns a gtsummary object of the results as below. This could rely on the get_cox_pairwise_df() function for computation and simply work on formatting these results into a table? Or it could compute the results itself if that is easier.

Example

library(NEST.TLGs)
library(tidyverse)
adam_db <- syntheticadam::get_data("adtte")

# data pre-processing ----------------------------------------------------------

arm_var <- "TRT01P"

df_ttet01 <- adam_db$adtte |>
  dplyr::filter(PARAMCD == "PFS" & ANL01FL == "Y") |>
  mutate(is_event = CNSR == 0,
         is_not_event = CNSR == 1,
         first_event = ifelse(CNSR == 0, EVNTDESC, NA_character_))

# Set the treatment arm variable as a factor with levels in desired order
df_ttet01[[arm_var]] <- factor(df_ttet01[[arm_var]], levels = c(
  "Placebo",
  "Xanomeline Low Dose",
  "Xanomeline High Dose"
))

# Build Cox PH Model --------------------------------------------------------------------

coxph_formula <- as.formula(sprintf("Surv(AVAL, is_event) ~ %s", arm_var))

df_coxph <- df_ttet01 |>
  get_cox_pairwise_df(
    model_formula = coxph_formula,
    arm = arm_var
    )

As you can see, this code generates the necessary statistics but does not format this in a gtsummary object.

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.