如何在 tbl_regression (gtsummary) 中添加 Cox 比例风险 model 的随访时间/人年？

Question

I use tbl_regression from the gtsummary package to depict a Cox proportional hazards model in a table and want to add follow-up time/person years to my tbl_regression table, but cannot seem to find out how to do this.

Can anyone help? Thank you!

Best, Mathilde

cox_cat_noncns <- coxph(Surv(TTD, Dod_status) ~ Highest_Edu_Household, data = data_cox_cat_noncns)

cox_cat_noncns_udj_table <- tbl_regression(cox_cat_noncns,
                                    label = Highest_Edu_Household ~ "Highest parental education",
                                    exponentiate = TRUE %>%
          add_nevent(location = "level") %>%
          bold_labels() %>%
          italicize_levels() %>%
          modify_tabel_styling(
             columns = estimate,
             rows = reference_row %in% TRUE,
             missing_symbol = "Ref.") %>%
          modify_footnote(everything() ~ NA, abbreviation = TRUE) %>%
          modify_table_styling(
             column = p.value,
             hide = TRUE) %>%
          modify_header(
             label= "")

Answer 1

There is a hidden column in the table with the total exposure time. The column is hidden by default, and you can unhide it using the modify_header() function (columns with assigned headers are unhidden). To see all columns available, inspect the tbl$table_body data frame. Happy Porgramming!

library(gtsummary)
library(survival)
packageVersion("gtsummary")
#> [1] '1.5.2'

tbl <-
  coxph(Surv(ttdeath, death) ~ age, trial) %>%
  tbl_regression(exponentiate = TRUE) %>%
  modify_header(exposure ~ "**Exposure**")

^{Created on 2022-04-17 by the reprex package (v2.0.1)}