Merge tbl_hierarchical() and tbl_hierarchical_count()

[p-lemercier]

Hi,
I want to generate hierarchical tables with both adverse event incidence rates and counts. Thus I tried to merge the results of tbl_hierarchical() and  tbl_hierarchical_count() but the resulting table is no longer hierachically ordered.

See examples below:

library(gtsummary)

ADAE_subset <- cards::ADAE |>
  dplyr::filter(
    AESOC %in% unique(cards::ADAE$AESOC)[1:5],
    AETERM %in% unique(cards::ADAE$AETERM)[1:5]
  )

# Event Rates --------------------
evt_rate <- tbl_hierarchical(
  data = ADAE_subset,
  variables = c(AESOC, AETERM),
  by = TRTA,
  denominator = cards::ADSL |> mutate(TRTA = ARM),
  id = USUBJID
)

# Event Counts -------------------
evt_cts <- tbl_hierarchical_count(
  data = ADAE_subset,
  variables = c(AESOC, AETERM),
  by = TRTA
)

evt_merge <- tbl_merge(
  list(evt_rate, evt_cts),
  tab_spanner = c("Rates", "Counts")
  )

evt_merge

I managed to reorder it manually this way :

evt_merge$table_body <- evt_merge$table_body |>
  dplyr::arrange(factor(label, levels = evt_rate$table_body$label))

evt_merge

I think tbl_merge should be able to manage hierarchical tables.

Thanks!

[ddsjoberg]

For sure, we'll get to this soon. Thanks for the post @p-lemercier !

[ddsjoberg]

NOTE TO SELF: We just need to update tbl_merge() to include grouping variables and levels in the variable-wise merging to retain the order. OR we could just recommend tbl_cbind() (#1855)