Gtsummary columns for all post hoc pairwise comparisons

Question

Is there currently a way in gtsummary to create columns for the p values of each pairwise comparison for a post hoc test? For example, say there are 4 groups, is there a function that includes the p values as separate columns for

1. the overall anova or chi square depending on if continuous or categorical
2. a vs b
3. a vs c
4. a vs d
5. b vs c
6. b vs d
7. c vs d

Answer 1

You can use the add_stat() function to add custom columns and rows to a tbl_summary() table. The example below uses the pairwise.t.test() to calculate the pairwise p-values (you can use whatever method you like).

library(gtsummary)
packageVersion("gtsummary")
#> [1] '1.5.0'
# set theme to get MEAN (SD) by default in `tbl_summary()`
theme_gtsummary_mean_sd()

# function to add pairwise copmarisons to `tbl_summary()`
add_stat_pairwise <- function(data, variable, by, ...) {
  # calculate pairwise p-values
  pw <- pairwise.t.test(data[[variable]], data[[by]], p.adj = "none")

  # convert p-values to list
  index <- 0L
  p.value.list <- list()
  for (i in seq_len(nrow(pw$p.value))) {
    for (j in seq_len(nrow(pw$p.value))) {
      index <- index + 1L
      
      p.value.list[[index]] <- 
        c(pw$p.value[i, j]) %>%
        setNames(glue::glue("**{colnames(pw$p.value)[j]} vs. {rownames(pw$p.value)[i]}**"))
    }
  }
  
  # convert list to data frame
  p.value.list %>% 
    unlist() %>%
    purrr::discard(is.na) %>%
    t() %>%
    as.data.frame() %>%
    # formatting/roundign p-values
    dplyr::mutate(dplyr::across(everything(), style_pvalue))
}

trial %>%
  select(grade, age, marker) %>%
  tbl_summary(by = grade, missing = "no") %>%
  # add pariwaise p-values
  add_stat(everything() ~ add_stat_pairwise) %>%
  as_kable() # convert to kable to display on Stackoverflow

Characteristic	I, N = 68	II, N = 68	III, N = 64	I vs. II	I vs. III	II vs. III
Age	46 (15)	48 (14)	48 (14)	0.6	0.4	0.8
Marker Level (ng/mL)	1.07 (0.89)	0.68 (0.73)	1.00 (0.92)	0.010	0.6	0.040

^{Created on 2021-11-26 by the reprex package (v2.0.1)}