FDA Table 33

Percentage of Patients Meeting Specific Hypotension Levels Postbaseline, Safety Population, Pooled Analysis

gtsummary Table Setup

# Load Libraries & Data
library(cardinal)

adsl <- random.cdisc.data::cadsl
advs <- random.cdisc.data::cadvs

# Output Table
make_table_33(df = advs, return_ard = FALSE)

Diastolic Blood Pressure (Pa)	A: Drug X N = 134¹	B: Placebo N = 134¹	C: Combination N = 132¹
SBP <90	9 (6.7%)	24 (17.9%)	12 (9.1%)
DBP <60	10 (7.5%)	13 (9.7%)	14 (10.6%)
¹ n (%)

Function Details

`make_table_33()`

df: USUBJID, AVISITN, PARAMCD, AVAL, AVALU, and the variables specified by arm_var and saffl_var.
denominator (if specified): USUBJID and the variables specified by arm_var and saffl_var.
If both the ARD (return_ard = "Y") and an overall column (lbl_overall not missing) are requested, a list of two ARDs will be returned: ard$tbl_summary for the table by groups, and ard$add_overall for the overal column.

Argument	Description	Default
`df`	(`data.frame`) Dataset (typically ADVS) required to build table.	No default
`denominator`	(`character`) Alternative dataset (typically ADSL) used only to calculate column counts.	`NULL`
`return_ard`	(`flag`) Whether an ARD should be returned.	`TRUE`
`id_vars`	(`character`) Identifier required to count the number of events (typically calculated by USUBJID).	`"USUBJID"`
`arm_var`	(`character`) Arm variable used to split table into columns.	`"ARM"`
`saffl_var`	(`character`) Flag variable used to indicate inclusion in safety population.	`"SAFFL"`
`lbl_overall`	(`character`) If specified, an overall column will be added to the table with the given value as the column label.	`NULL`
`subset`	(`character`) If specified, the calculations will be done on the selected subset of `df`.	`NULL`

Source code for this function is available here.

ARD Setup

# Load Libraries & Data
library(cardinal)

adsl <- random.cdisc.data::cadsl
advs <- random.cdisc.data::cadvs

# Create Table & ARD
result <- make_table_33(df = advs, return_ard = TRUE)

# Output ARD
result$ard

$tbl_summary
   group1 group1_level variable variable_level stat_name stat_label  stat
1     ARM    A: Drug X    SBP90           TRUE         n          n     9
2     ARM    A: Drug X    SBP90           TRUE         N          N   134
3     ARM    A: Drug X    SBP90           TRUE         p          % 0.067
4     ARM    A: Drug X    DBP60           TRUE         n          n    10
5     ARM    A: Drug X    DBP60           TRUE         N          N   134
6     ARM    A: Drug X    DBP60           TRUE         p          % 0.075
7     ARM    B: Place…    SBP90           TRUE         n          n    24
8     ARM    B: Place…    SBP90           TRUE         N          N   134
9     ARM    B: Place…    SBP90           TRUE         p          % 0.179
10    ARM    B: Place…    DBP60           TRUE         n          n    13

rtables Table Setup

# Load Libraries & Data
library(cardinal)

adsl <- random.cdisc.data::cadsl
advs <- random.cdisc.data::cadvs

# Output Table
risk_diff <- list(arm_x = "B: Placebo", arm_y = "A: Drug X") # optional
make_table_33_rtables(
  df = advs,
  alt_counts_df = adsl,
  risk_diff = risk_diff
)

Blood Pressure   A: Drug X   B: Placebo   C: Combination   Risk Difference (%) (95% CI)
(Pa)              (N=134)     (N=134)        (N=132)                 (N=268)           
———————————————————————————————————————————————————————————————————————————————————————
SBP <90          9 (6.7%)    24 (17.9%)     12 (9.1%)           11.2 (3.4 - 18.9)      
DBP <60          10 (7.5%)   13 (9.7%)      14 (10.6%)           2.2 (-4.5 - 8.9)

Function Details

`make_table_33_rtables()`

Required variables:

df: USUBJID, AVISITN, PARAMCD, AVAL, AVALU, and the variables specified by arm_var and saffl_var.
alt_counts_df (if specified): USUBJID and the variables specified by arm_var and saffl_var.

Argument	Description	Default
`df`	(`data.frame`) Dataset (typically ADVS) required to build table.	No default
`alt_counts_df`	(`character`) Alternative dataset (typically ADSL) used only to calculate column counts.	`NULL`
`show_colcounts`	(`flag`) Whether column counts should be printed.	`TRUE`
`arm_var`	(`character`) Arm variable used to split table into columns.	`"ARM"`
`saffl_var`	(`character`) Flag variable used to indicate inclusion in safety population.	`"SAFFL"`
`lbl_overall`	(`character`) If specified, an overall column will be added to the table with the given value as the column label.	`NULL`
`risk_diff`	(named `list`) List of settings to apply to add a risk difference column to the table. See tern::add_riskdiff() for more details. List should contain the following elements: `arm_x`: (required) the name of reference arm. `arm_y`: (required) the name of the arm to compare to the reference arm. `col_label`: (optional) label to use for the risk difference column. Defaults to `"Risk Difference (%) (95% CI)"`. `pct`: (optional) whether the output should be returned as percentages. Defaults to `TRUE`.	`NULL`
`prune_0`	(`flag`) Whether all-zero rows should be removed from the table.	`FALSE`
`annotations`	(named `list` of `character`) List of annotations to add to the table. Valid annotation types are `title`, `subtitles`, `main_footer`, and `prov_footer.` Each name-value pair should use the annotation type as name and the desired string as value.	`NULL`

Source code for this function is available here.

--- title: FDA Table 33 subtitle: Percentage of Patients Meeting Specific Hypotension Levels Postbaseline, Safety Population, Pooled Analysis format: html --- ::: panel-tabset ## Spec. Screenshot ![](../assets/images/screenshots/table_33.png){fig-align="center" width="70%"} ## gtsummary Table <details> <summary>gtsummary Table Setup</summary> ```{r tbl1, eval=FALSE, echo=TRUE} # Load Libraries & Data library(cardinal) adsl <- random.cdisc.data::cadsl advs <- random.cdisc.data::cadvs # Output Table make_table_33(df = advs, return_ard = FALSE) ``` </details> ```{r tbl1, message=FALSE, warning=FALSE, eval=TRUE} ``` <details> <summary>Function Details</summary> ### `make_table_33()` - **`df`**: `USUBJID`, `AVISITN`, `PARAMCD`, `AVAL`, `AVALU`, and the variables specified by `arm_var` and `saffl_var`. - **`denominator`** (if specified): `USUBJID` and the variables specified by `arm_var` and `saffl_var`. - If both the ARD (`return_ard = "Y"`) and an overall column (`lbl_overall` not missing) are requested, a list of two ARDs will be returned: `ard$tbl_summary` for the table by groups, and `ard$add_overall` for the overal column. +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | **Argument** | **Description** | **Default** | +:==============+:===================================================================================================================+:=============+ | `df` | (`data.frame`) Dataset (typically ADVS) required to build table. | *No default* | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `denominator` | (`character`) Alternative dataset (typically ADSL) used only to calculate column counts. | `NULL` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `return_ard` | (`flag`) Whether an ARD should be returned. | `TRUE` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `id_vars` | (`character`) Identifier required to count the number of events (typically calculated by USUBJID). | `"USUBJID"` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `arm_var` | (`character`) Arm variable used to split table into columns. | `"ARM"` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `saffl_var` | (`character`) Flag variable used to indicate inclusion in safety population. | `"SAFFL"` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `lbl_overall` | (`character`) If specified, an overall column will be added to the table with the given value as the column label. | `NULL` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ | `subset` | (`character`) If specified, the calculations will be done on the selected subset of `df`. | `NULL` | +---------------+--------------------------------------------------------------------------------------------------------------------+--------------+ Source code for this function is available [here](https://github.com/pharmaverse/cardinal/blob/main/R/fda-table_05.R). ------------------------------------------------------------------------ </details> ## ARD <details> <summary>ARD Setup</summary> ```{r tbl2, eval=FALSE, echo=TRUE} # Load Libraries & Data library(cardinal) adsl <- random.cdisc.data::cadsl advs <- random.cdisc.data::cadvs # Create Table & ARD result <- make_table_33(df = advs, return_ard = TRUE) # Output ARD result$ard ``` </details> ```{r tbl2, message=FALSE, warning=FALSE, eval=TRUE} ``` ## rtables Table <details> <summary>rtables Table Setup</summary> ```{r tbl3, eval=FALSE, echo=TRUE} # Load Libraries & Data library(cardinal) adsl <- random.cdisc.data::cadsl advs <- random.cdisc.data::cadvs # Output Table risk_diff <- list(arm_x = "B: Placebo", arm_y = "A: Drug X") # optional make_table_33_rtables( df = advs, alt_counts_df = adsl, risk_diff = risk_diff ) ``` </details> ```{r tbl3, message=FALSE, warning=FALSE, eval=TRUE} ``` <details> <summary>Function Details</summary> ### `make_table_33_rtables()` ------------------------------------------------------------------------ Required variables: - **`df`**: `USUBJID`, `AVISITN`, `PARAMCD`, `AVAL`, `AVALU`, and the variables specified by `arm_var` and `saffl_var`. - **`alt_counts_df`** (if specified): `USUBJID` and the variables specified by `arm_var` and `saffl_var`. +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | Argument | Description | Default | +:=================+:=============================================================================================================================================================================================================================================================+:=============+ | `df` | (`data.frame`) Dataset (typically ADVS) required to build table. | *No default* | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `alt_counts_df` | (`character`) Alternative dataset (typically ADSL) used only to calculate column counts. | `NULL` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `show_colcounts` | (`flag`) Whether column counts should be printed. | `TRUE` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `arm_var` | (`character`) Arm variable used to split table into columns. | `"ARM"` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `saffl_var` | (`character`) Flag variable used to indicate inclusion in safety population. | `"SAFFL"` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `lbl_overall` | (`character`) If specified, an overall column will be added to the table with the given value as the column label. | `NULL` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `risk_diff` | (named `list`) List of settings to apply to add a risk difference column to the table. See [tern::add_riskdiff()](https://insightsengineering.github.io/tern/main/reference/add_riskdiff.html) for more details. List should contain the following elements: | `NULL` | | | | | | | - `arm_x`: (required) the name of reference arm. | | | | | | | | - `arm_y`: (required) the name of the arm to compare to the reference arm. | | | | | | | | - `col_label`: (optional) label to use for the risk difference column. Defaults to `"Risk Difference (%) (95% CI)"`. | | | | | | | | - `pct`: (optional) whether the output should be returned as percentages. Defaults to `TRUE`. | | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `prune_0` | (`flag`) Whether all-zero rows should be removed from the table. | `FALSE` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ | `annotations` | (named `list` of `character`) List of annotations to add to the table. Valid annotation types are `title`, `subtitles`, `main_footer`, and `prov_footer.` Each name-value pair should use the annotation type as name and the desired string as value. | `NULL` | +------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--------------+ Source code for this function is available [here](https://github.com/pharmaverse/falcon/blob/main/R/fda-table_33.R). </details> :::