Add `tbl_heirarchical()` #1872

ddsjoberg · 2024-07-25T23:32:07Z

We need an ARD-based function for items like AE summaries. This still needs some thought.

First proposed API....

tbl_hierarchical(
  data = ADAE, 
  heirarchy = c(SOC, AE),
  by, # can we use a single argument for stratifying tables by treatment and AE grade, for example?
  statistic = "{n} ({p})",
  denominator = ADSL, 
  id = USUBJID,
  include = everything(), # this would be the variables from `heirarchy` that we would include summary stats for (some of the nested drug class tables don't need stats on the class level)
  overall_row = FALSE
)

Would it work if the the first by variable listed was calculated as is, then any subsequent by variables are processed with tbl_strata()? e.g. if by=ARM, then we would have two columns of statistics (treatment A, treatment B). If by=c(ARM, AEGRADE), then the first header row would be the 5 grade levels, then the header row above that would be for treatments A and B... 🤔

The text was updated successfully, but these errors were encountered:

ddsjoberg · 2024-09-07T23:12:13Z

Aiming to use this function to build tables from the FDA specs

edelarua · 2024-10-03T21:19:16Z

FDA Table 9:

# Load Libraries & Data

library(cardinal)
library(gtsummary)

adsl <- random.cdisc.data::cadsl
adae <- random.cdisc.data::cadae

# cardinal

make_table_09(adae = adae, alt_counts_df = adsl, pref_var = "AEDECOD")
#> System Organ Class           A: Drug X    B: Placebo    C: Combination
#>   Dictionary-Derived Term     (N=134)       (N=134)        (N=132)    
#> ——————————————————————————————————————————————————————————————————————
#> Any SAE                     104 (77.6%)   101 (75.4%)     99 (75.0%)  
#> cl A                        48 (35.8%)    48 (35.8%)      50 (37.9%)  
#>   dcd A.1.1.1.2             48 (35.8%)    48 (35.8%)      50 (37.9%)  
#> cl B                        79 (59.0%)    78 (58.2%)      76 (57.6%)  
#>   dcd B.1.1.1.1             47 (35.1%)    49 (36.6%)      43 (32.6%)  
#>   dcd B.2.2.3.1             48 (35.8%)    54 (40.3%)      51 (38.6%)  
#> cl D                        50 (37.3%)    42 (31.3%)      51 (38.6%)  
#>   dcd D.1.1.1.1             50 (37.3%)    42 (31.3%)      51 (38.6%)

# gtsummary

adae <- adae %>%
  dplyr::filter(SAFFL == "Y", AESER == "Y")

tbl_hierarchical(
  adae,
  by = ARM,
  variables = c(AESOC, AEDECOD),
  id = USUBJID,
  denominator = adsl,
  overall_row = TRUE,
  label = list(AESOC = "System Organ Class", overall = "Any SAE")
)

^{Created on 2024-10-03 with reprex v2.1.1}

edelarua · 2024-10-03T21:21:57Z

FDA Table 10:

# Load Libraries & Data

library(dplyr)
library(cardinal)
library(gtsummary)

adsl <- random.cdisc.data::cadsl
adae <- random.cdisc.data::cadae

# Pre-Processing - Ensure required variables fmqsc_var and fmqnam_var exist in adae
set.seed(1)
adae <- adae %>%
  rename(FMQ01SC = SMQ01SC) %>%
  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"

# cardinal

make_table_10(adae = adae, alt_counts_df = adsl)
#> Body System or Organ Class   A: Drug X    B: Placebo   C: Combination
#>   FMQ (Narrow)                (N=134)      (N=134)        (N=132)    
#> —————————————————————————————————————————————————————————————————————
#> cl A.1                                                               
#>   FMQ1                       17 (12.7%)   14 (10.4%)     29 (22.0%)  
#>   FMQ2                       23 (17.2%)   20 (14.9%)     20 (15.2%)  
#>   FMQ3                       20 (14.9%)   19 (14.2%)     23 (17.4%)  
#> cl B.1                                                               
#>   FMQ1                        8 (6.0%)    11 (8.2%)       7 (5.3%)   
#>   FMQ2                        5 (3.7%)    12 (9.0%)      16 (12.1%)  
#>   FMQ3                       10 (7.5%)     7 (5.2%)       5 (3.8%)   
#> cl B.2                                                               
#>   FMQ1                       13 (9.7%)    15 (11.2%)      9 (6.8%)   
#>   FMQ2                       12 (9.0%)     9 (6.7%)      10 (7.6%)   
#>   FMQ3                        6 (4.5%)     7 (5.2%)       9 (6.8%)   
#> cl C.2                                                               
#>   FMQ1                        9 (6.7%)     9 (6.7%)      12 (9.1%)   
#>   FMQ2                        6 (4.5%)     8 (6.0%)       8 (6.1%)   
#>   FMQ3                        6 (4.5%)     9 (6.7%)      10 (7.6%)   
#> cl D.1                                                               
#>   FMQ1                       23 (17.2%)   17 (12.7%)     27 (20.5%)  
#>   FMQ2                       22 (16.4%)   20 (14.9%)     25 (18.9%)  
#>   FMQ3                       15 (11.2%)   19 (14.2%)     21 (15.9%)  
#> cl D.2                                                               
#>   FMQ1                        8 (6.0%)    11 (8.2%)       9 (6.8%)   
#>   FMQ2                       14 (10.4%)   15 (11.2%)     14 (10.6%)  
#>   FMQ3                       11 (8.2%)     9 (6.7%)      11 (8.3%)

# gtsummary

adae <- adae %>%
  filter(SAFFL == "Y", AESER == "Y", FMQ01SC == "NARROW")

tbl_hierarchical(
  adae,
  by = ARM,
  variables = c(AEBODSYS, FMQ01NAM),
  include = FMQ01NAM,
  id = USUBJID,
  denominator = adsl,
  label = list(FMQ01NAM = "FMQ (Narrow)")
)

^{Created on 2024-10-03 with reprex v2.1.1}

edelarua · 2024-10-03T21:29:29Z

AET02 (simplified):

# Load Libraries & Data

library(dplyr)
library(tern)
library(gtsummary)

adsl <- random.cdisc.data::cadsl
adae <- random.cdisc.data::cadae

adsl <- df_explicit_na(adsl)
adae <- df_explicit_na(adae) %>%
  var_relabel(
    AEBODSYS = "MedDRA System Organ Class",
    AEDECOD = "MedDRA Preferred Term"
  ) %>%
  filter(ANL01FL == "Y")

# rtables/tern

split_fun <- drop_split_levels

lyt <- basic_table(show_colcounts = TRUE) %>%
  split_cols_by(var = "ACTARM") %>%
  analyze_num_patients(
    vars = "USUBJID",
    .stats = "unique",
    .labels = c(
      unique = "Total number of patients with at least one adverse event"
    )
  ) %>%
  split_rows_by(
    "AEBODSYS",
    child_labels = "visible",
    nested = FALSE,
    split_fun = split_fun,
    label_pos = "topleft",
    split_label = obj_label(adae$AEBODSYS)
  ) %>%
  count_occurrences(
    vars = "AEDECOD"
  ) %>%
  append_varlabels(adae, "AEDECOD", indent = 1L)

result <- build_table(lyt, df = adae, alt_counts_df = adsl)
result
#> MedDRA System Organ Class                                   A: Drug X    B: Placebo   C: Combination
#>   MedDRA Preferred Term                                      (N=134)      (N=134)        (N=132)    
#> ————————————————————————————————————————————————————————————————————————————————————————————————————
#> Total number of patients with at least one adverse event   100 (74.6%)   98 (73.1%)    103 (78.0%)  
#> cl A.1                                                                                              
#>   dcd A.1.1.1.1                                            45 (33.6%)    31 (23.1%)     52 (39.4%)  
#>   dcd A.1.1.1.2                                            41 (30.6%)    39 (29.1%)     42 (31.8%)  
#> cl B.1                                                                                              
#>   dcd B.1.1.1.1                                            38 (28.4%)    37 (27.6%)     36 (27.3%)  
#> cl B.2                                                                                              
#>   dcd B.2.1.2.1                                            39 (29.1%)    34 (25.4%)     46 (34.8%)  
#>   dcd B.2.2.3.1                                            38 (28.4%)    40 (29.9%)     45 (34.1%)  
#> cl C.1                                                                                              
#>   dcd C.1.1.1.3                                            36 (26.9%)    34 (25.4%)     36 (27.3%)  
#> cl C.2                                                                                              
#>   dcd C.2.1.2.1                                            28 (20.9%)    36 (26.9%)     48 (36.4%)  
#> cl D.1                                                                                              
#>   dcd D.1.1.1.1                                            42 (31.3%)    32 (23.9%)     46 (34.8%)  
#>   dcd D.1.1.4.2                                            38 (28.4%)    34 (25.4%)     40 (30.3%)  
#> cl D.2                                                                                              
#>   dcd D.2.1.5.3                                            37 (27.6%)    46 (34.3%)     50 (37.9%)

# gtsummary

tbl_hierarchical(
  adae,
  by = ACTARM,
  variables = c(AEBODSYS, AEDECOD),
  id = USUBJID,
  denominator = adsl,
  include = AEDECOD,
  overall_row = TRUE,
  label = list(overall = "Total number of patients with at least one adverse event")
)

^{Created on 2024-10-03 with reprex v2.1.1}

edelarua · 2024-10-03T22:35:42Z

AET04 (simplified):

# Load Libraries & Data

library(dplyr)
library(tern)
library(gtsummary)

adsl <- random.cdisc.data::cadsl
adae <- random.cdisc.data::cadae

adsl <- df_explicit_na(adsl) %>% filter(TRT01A != "<Missing>")
adae <- df_explicit_na(adae) %>%
  var_relabel(
    AEBODSYS = "MedDRA System Organ Class",
    AEDECOD = "MedDRA Preferred Term"
  ) %>%
  filter(
    ANL01FL == "Y",
    AETOXGR != "<Missing>"
  )

# rtables/tern

lyt <- basic_table(show_colcounts = TRUE) %>%
  split_cols_by("ACTARM") %>%
  split_rows_by(
    "AEBODSYS",
    child_labels = "visible",
    nested = FALSE,
    split_fun = drop_split_levels,
    split_label = var_labels(adae)[["AEBODSYS"]],
    label_pos = "topleft"
  ) %>%
  count_occurrences_by_grade(
    var = "AETOXGR"
  ) %>%
  append_topleft("  Highest Grade")

result <- lyt %>%
  build_table(adae, alt_counts_df = adsl) %>%
  prune_table()

result
#> MedDRA System Organ Class   A: Drug X    B: Placebo   C: Combination
#>   Highest Grade              (N=134)      (N=134)        (N=132)    
#> ————————————————————————————————————————————————————————————————————
#> cl A.1                                                              
#>   1                         27 (20.1%)   19 (14.2%)     34 (25.8%)  
#>   2                         41 (30.6%)   39 (29.1%)     42 (31.8%)  
#> cl B.1                                                              
#>   5                         38 (28.4%)   37 (27.6%)     36 (27.3%)  
#> cl B.2                                                              
#>   1                         23 (17.2%)   22 (16.4%)     28 (21.2%)  
#>   3                         39 (29.1%)   34 (25.4%)     46 (34.8%)  
#> cl C.1                                                              
#>   4                         36 (26.9%)   34 (25.4%)     36 (27.3%)  
#> cl C.2                                                              
#>   2                         28 (20.9%)   36 (26.9%)     48 (36.4%)  
#> cl D.1                                                              
#>   3                         22 (16.4%)   22 (16.4%)     22 (16.7%)  
#>   5                         42 (31.3%)   32 (23.9%)     46 (34.8%)  
#> cl D.2                                                              
#>   1                         37 (27.6%)   46 (34.3%)     50 (37.9%)

# gtsummary

tbl_hierarchical(
  adae |> mutate(AETOXGR = factor(AETOXGR, ordered = TRUE)),
  by = ACTARM,
  variables = c(AEBODSYS, AETOXGR),
  id = USUBJID,
  denominator = adsl,
  include = AETOXGR,
  label = AETOXGR ~ "Highest Grade"
) |>
  # prune empty rows
  modify_table_body(
    ~ .x |>
      filter(
        rowSums(.x[c("stat_1", "stat_2", "stat_3")] != "0 (0.0)") > 0 |
          is.na(.x$stat_1)
      )
  )
#> ℹ Denominator set by "ACTARM" column in `denominator` data frame.

^{Created on 2024-10-03 with reprex v2.1.1}

ddsjoberg added this to the v2.1.0 milestone Jul 25, 2024

ddsjoberg assigned edelarua Sep 9, 2024

ddsjoberg mentioned this issue Sep 23, 2024

Add ARD-first example to pharmaverse examples page #1949

Open

edelarua mentioned this issue Oct 1, 2024

Add tbl_hierarchical() #2013

Merged

11 tasks

edelarua closed this as completed in #2013 Oct 3, 2024

Melkiades mentioned this issue Dec 27, 2024

[investigation] provide list of edge cases for customized tables insightsengineering/rtables#952

Closed

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add `tbl_heirarchical()` #1872

Add `tbl_heirarchical()` #1872

ddsjoberg commented Jul 25, 2024 •

edited

Loading

ddsjoberg commented Sep 7, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

Add tbl_heirarchical() #1872

Add tbl_heirarchical() #1872

Comments

ddsjoberg commented Jul 25, 2024 • edited Loading

ddsjoberg commented Sep 7, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

edelarua commented Oct 3, 2024

Add `tbl_heirarchical()` #1872

Add `tbl_heirarchical()` #1872

ddsjoberg commented Jul 25, 2024 •

edited

Loading