How to add two difference statistics in tbl_summary? - gtsummary

Question

I am a beginner in r and trying to create some tables using this great gtsummary package. My question is that is it possible to use add_difference to add two separate difference statistics at the same time (or to combine these somehow)? I am able to create a perfect table with p-values or effect size, but not with both. Also, is it possible to use bonferroni adjusted p-values?
My simple code (with t.test) looks like this:
table1 <- tbl_summary(df, by = gr, statistic = list(all_continuous() ~ "{mean} ({sd})")) %>% add_difference(test = list(all_continuous() ~ "t.test"), group = gr, conf.level = 0.95, pvalue_fun = function(x) style_pvalue(x, digits = 2)))
thanks for help.

Answer 1

Yes, the table you're after is possible in gtsummary. Use the tbl_merge() function to merge the tables with standardized differences and the p-values. There is an example of this below. An alternative to this is to use add_stat() to construct customized columns.
library(gtsummary)
#> #BlackLivesMatter
packageVersion("gtsummary")
#> [1] '1.4.1'
# standardized difference
tbl1 <-
trial %>%
select(trt, age, marker) %>%
tbl_summary(by = trt, missing = "no",
statistic = all_continuous() ~ "{mean} ({sd})") %>%
add_difference(all_continuous() ~ "cohens_d")
# table with p-value and corrected p-values
tbl2 <-
trial %>%
select(trt, age, marker) %>%
tbl_summary(by = trt, missing = "no") %>%
add_p(all_continuous() ~ "t.test") %>%
add_q(method = "bonferroni") %>%
# hide the trt summaries, because we don't need them repeated
modify_column_hide(all_stat_cols())
#> add_q: Adjusting p-values with
#> `stats::p.adjust(x$table_body$p.value, method = "bonferroni")`
# merge tbls together
tbl_final <-
tbl_merge(list(tbl1, tbl2)) %>%
# remove spanning headers
modify_spanning_header(everything() ~ NA)
Created on 2021-07-20 by the reprex package (v2.0.0)