make_table_10_gt

[YoshitoKoujin]

-update argument for AESOC specification

Pull Request

Closes #nnn

[github-actions]

Code Coverage Summary

Filename            Stmts    Miss  Cover    Missing
----------------  -------  ------  -------  ---------
R/fda-fig_01.R         75       0  100.00%
R/fda-fig_02.R         75       0  100.00%
R/fda-fig_14.R        115       0  100.00%
R/fda-table_02.R      166       0  100.00%
R/fda-table_03.R       35       0  100.00%
R/fda-table_04.R       63       0  100.00%
R/fda-table_05.R       61       0  100.00%
R/fda-table_06.R       76       0  100.00%
R/fda-table_07.R       35       0  100.00%
R/fda-table_08.R       46       0  100.00%
R/fda-table_09.R      333       4  98.80%   254-258
R/fda-table_10.R       98       1  98.98%   194
R/fda-table_11.R       42       0  100.00%
R/fda-table_12.R       40       0  100.00%
R/fda-table_13.R       20       0  100.00%
R/fda-table_14.R       27       0  100.00%
R/fda-table_15.R       32       0  100.00%
R/fda-table_16.R       32       0  100.00%
R/fda-table_17.R       32       0  100.00%
R/fda-table_18.R       32       0  100.00%
R/fda-table_20.R       71       0  100.00%
R/fda-table_21.R       78       0  100.00%
R/fda-table_22.R       77       0  100.00%
R/fda-table_32.R       94       0  100.00%
R/fda-table_33.R       95       0  100.00%
R/fda-table_34.R       30       0  100.00%
R/fda-table_35.R       23       1  95.65%   58
R/fda-table_36.R       30       1  96.67%   70
R/fda-table_38.R       39       0  100.00%
R/utils.R              28       0  100.00%
TOTAL                2000       7  99.65%

Diff against main

Filename            Stmts    Miss  Cover
----------------  -------  ------  -------
R/fda-table_10.R      +72      +1  -1.02%
TOTAL                 +72      +1  -0.04%

Results for commit: 09e4b51

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
fda-table_10	💔	$1.20$	$+19.52$	$+8$	$+2$	$0$	$0$

Additional test case details

Test Suite	$Status$	Time on main	$±Time$	Test Case
fda-table_10	👶		$+13.25$	Table_10_gt_generation_works
fda-table_10	👶		$+6.26$	Table_10_gt_generation_works_with_custom_values

Results for commit 15b5d97

♻️ This comment has been updated with latest results.

[ayogasekaram]

Thank you so much for working on this @YoshitoKoujin! I have a couple of suggestions:

1. I'm looking at the FDA spec file and it looks like there are some rows in the gt table that aren't in the spec -

Is there a way to remove the numbers from these rows?

2. is there a way to compute this table using gt_sum functions? with the gt table, I believe it's not only producing a gt end product but also using their functions to calculate the values. maybe you can consult @ddsjoberg on a direction for this table.

[ddsjoberg]

There is a gtsummary extension that makes these types of tables (kind of, with a little manipulation). In the next release of gtsummary, these calculations will be 1. More straight-forward, 2. Much, much faster.

adsl <- scda::synthetic_cdisc_dataset("rcd_2022_10_13", "adsl")
adae <- scda::synthetic_cdisc_dataset("rcd_2022_10_13", "adae")

# Pre-Processing - Ensure required variables fmqsc_var and fmqnam_var exist in adae
set.seed(1)
adae <- adae |> 
  dplyr::rename(FMQ01SC = SMQ01SC) |> 
  dplyr::mutate(
    AESER = sample(c("Y", "N"), size = nrow(adae), replace = TRUE),
    FMQ01NAM = sample(c("FMQ1", "FMQ2", "FMQ3"), size = nrow(adae), replace = TRUE)
  )
adae$FMQ01SC[is.na(adae$FMQ01SC)] <- "NARROW"

example_table <- 
  # build primary table
  gtsummary::tbl_strata2(
    data = adae,
    strata = "AEBODSYS",
    ~gtreg::tbl_ae(
      data = .x |> dplyr::mutate(.x, AEBODSYS = .y),
      strata = "ARM",
      id = "USUBJID",
      soc = "AEBODSYS",
      ae = "FMQ01NAM",
      statistic = "{n} ({p}%)",
      id_df = adsl
    ),
    .combine_with = "tbl_stack",
    .combine_args = list(group_header = NULL)
  ) |> 
  # remove stats from SOC row
  gtsummary::modify_table_body(
    ~.x |> 
      dplyr::mutate(
        dplyr::across(
          gtsummary::all_stat_cols(),
          function(x) ifelse(.data$variable %in% "soc", "", x)
        )
      )
  ) |> 
  # update column headers
  gtsummary::modify_header(
    gtreg::all_ae_cols() ~ "**{strata}**  \nN = {n}",
    label = "**Body System or Organ Class**  \nFMQ (Narrow)"
  ) |> 
  gtsummary::modify_spanning_header(gtreg::all_ae_cols() ~ NA) |> 
  # bold SOC rows
  gtsummary::modify_table_styling(
    columns = "label",
    rows = variable %in% "soc",
    text_format = "bold"
  ) 

[YoshitoKoujin]

There is a gtsummary extension that makes these types of tables (kind of, with a little manipulation). In the next release of gtsummary, these calculations will be 1. More straight-forward, 2. Much, much faster.

adsl <- scda::synthetic_cdisc_dataset("rcd_2022_10_13", "adsl")
adae <- scda::synthetic_cdisc_dataset("rcd_2022_10_13", "adae")

# Pre-Processing - Ensure required variables fmqsc_var and fmqnam_var exist in adae
set.seed(1)
adae <- adae |> 
  dplyr::rename(FMQ01SC = SMQ01SC) |> 
  dplyr::mutate(
    AESER = sample(c("Y", "N"), size = nrow(adae), replace = TRUE),
    FMQ01NAM = sample(c("FMQ1", "FMQ2", "FMQ3"), size = nrow(adae), replace = TRUE)
  )
adae$FMQ01SC[is.na(adae$FMQ01SC)] <- "NARROW"

example_table <- 
  # build primary table
  gtsummary::tbl_strata2(
    data = adae,
    strata = "AEBODSYS",
    ~gtreg::tbl_ae(
      data = .x |> dplyr::mutate(.x, AEBODSYS = .y),
      strata = "ARM",
      id = "USUBJID",
      soc = "AEBODSYS",
      ae = "FMQ01NAM",
      statistic = "{n} ({p}%)",
      id_df = adsl
    ),
    .combine_with = "tbl_stack",
    .combine_args = list(group_header = NULL)
  ) |> 
  # remove stats from SOC row
  gtsummary::modify_table_body(
    ~.x |> 
      dplyr::mutate(
        dplyr::across(
          gtsummary::all_stat_cols(),
          function(x) ifelse(.data$variable %in% "soc", "", x)
        )
      )
  ) |> 
  # update column headers
  gtsummary::modify_header(
    gtreg::all_ae_cols() ~ "**{strata}**  \nN = {n}",
    label = "**Body System or Organ Class**  \nFMQ (Narrow)"
  ) |> 
  gtsummary::modify_spanning_header(gtreg::all_ae_cols() ~ NA) |> 
  # bold SOC rows
  gtsummary::modify_table_styling(
    columns = "label",
    rows = variable %in% "soc",
    text_format = "bold"
  ) 

Thank you for the info. I'm looking forward to the next release!

[YoshitoKoujin]

Thank you for your review @ayogasekaram! I've made some adjustments and removed a few rows based on your comments. Could you please take a look at the updated program? This version utilizes 'gt', but does not incorporate 'gtsummary'. However, we will need to update it once the new 'gtsummary' is released. Therefore, please consider this program as a provisional version.

[github-actions]

Unit Tests Summary

  1 files   29 suites   1m 31s ⏱️
130 tests 130 ✅   0 💤 0 ❌
473 runs  351 ✅ 122 💤 0 ❌

Results for commit 09e4b51.

[ayogasekaram]

This PR will be revisited when the gtsummary package updates are complete and released.