Sort Stacked Hierarchical ARDs

This function is used to sort stacked hierarchical ARDs.

For the purposes of this function, we define a "variable group" as a combination of ARD rows grouped by the combination of all their variable levels, but excluding any by variables.

sort_ard_hierarchical(x, sort = everything() ~ "descending", by_level = NULL)

Arguments

x

(card)
a stacked hierarchical ARD of class 'card' created using ard_stack_hierarchical() or ard_stack_hierarchical_count().

sort

(formula-list-selector, string)
a named list, a list of formulas, a single formula where the list element is a named list of functions (or the RHS of a formula), or a single string specifying the types of sorting to perform at each hierarchy variable level. If the sort method for any variable is not specified then the method will default to "descending". If a single unnamed string is supplied it is applied to all variables. For each variable, the value specified must be one of:

"alphanumeric" - at the specified hierarchy level of the ARD, groups are ordered alphanumerically (i.e. A to Z) by variable_level text.
"descending" - within each variable group of the ARD at the specified hierarchy level, count sums are calculated for each group and groups are sorted in descending order by sum. When sort is "descending" for a given variable and n is included in statistic for the variable then n is used to calculate variable group sums, otherwise p is used. If neither n nor p are present in x for the variable, an error will occur.

Defaults to everything() ~ "descending".

by_level

(named list)
a named list used to restrict the counts used for "descending" sorting to one or more specific by variable levels. Each name must be one of the by variables used to create x, and each element must be a single level of that by variable. When supplied, "descending" sorts rank variable groups by the count sums calculated only from the rows matching every specified by variable level (e.g. list(TRTA = "Placebo") sorts by the counts observed in the "Placebo" arm), rather than the sums across all by variable levels. Any by variable not named in the list is not restricted. This argument has no effect on variables sorted "alphanumeric".

Defaults to NULL, in which case count sums are calculated across all by variable levels.

Value

an ARD data frame of class 'card'

Note

If overall data is present in x (i.e. the ARD was created with ard_stack_hierarchical(overall=TRUE)), the overall data will be sorted last within each variable group (i.e. after any other rows with the same combination of variable levels).

Examples

ard_stack_hierarchical(
  ADAE,
  variables = c(AESOC, AEDECOD),
  by = TRTA,
  denominator = ADSL,
  id = USUBJID
) |>
  sort_ard_hierarchical(AESOC ~ "alphanumeric")
#> # An ARD data frame: 2,394 × 13
#>    group1 group1_level group2 group2_level variable variable_level     stat_name
#>    <chr>  <list>       <chr>  <list>       <chr>    <list>             <chr>    
#>  1 NA     <NULL>       NA     <NULL>       TRTA     Placebo            n        
#>  2 NA     <NULL>       NA     <NULL>       TRTA     Placebo            N        
#>  3 NA     <NULL>       NA     <NULL>       TRTA     Placebo            p        
#>  4 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… n        
#>  5 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… N        
#>  6 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… p        
#>  7 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… n        
#>  8 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… N        
#>  9 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… p        
#> 10 TRTA   Placebo      NA     <NULL>       AESOC    CARDIAC DISORDERS  n        
#> # ℹ 2,384 more rows
#> # ℹ 6 more variables: context <chr>, stat_label <chr>, stat <list>,
#> #   fmt_fun <list>, warning <list>, error <list>

ard_stack_hierarchical_count(
  ADAE,
  variables = c(AESOC, AEDECOD),
  by = TRTA,
  denominator = ADSL
) |>
  sort_ard_hierarchical(sort = list(AESOC ~ "alphanumeric", AEDECOD ~ "descending"))
#> # An ARD data frame: 804 × 13
#>    group1 group1_level group2 group2_level variable variable_level     stat_name
#>    <chr>  <list>       <chr>  <list>       <chr>    <list>             <chr>    
#>  1 NA     <NULL>       NA     <NULL>       TRTA     Placebo            n        
#>  2 NA     <NULL>       NA     <NULL>       TRTA     Placebo            N        
#>  3 NA     <NULL>       NA     <NULL>       TRTA     Placebo            p        
#>  4 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… n        
#>  5 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… N        
#>  6 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… p        
#>  7 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… n        
#>  8 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… N        
#>  9 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… p        
#> 10 TRTA   Placebo      NA     <NULL>       AESOC    CARDIAC DISORDERS  n        
#> # ℹ 794 more rows
#> # ℹ 6 more variables: context <chr>, stat_label <chr>, stat <list>,
#> #   fmt_fun <list>, warning <list>, error <list>

# sort by the counts observed in the "Placebo" arm only
ard_stack_hierarchical(
  ADAE,
  variables = c(AESOC, AEDECOD),
  by = TRTA,
  denominator = ADSL,
  id = USUBJID
) |>
  sort_ard_hierarchical(by_level = list(TRTA = "Placebo"))
#> # An ARD data frame: 2,394 × 13
#>    group1 group1_level group2 group2_level variable variable_level     stat_name
#>    <chr>  <list>       <chr>  <list>       <chr>    <list>             <chr>    
#>  1 NA     <NULL>       NA     <NULL>       TRTA     Placebo            n        
#>  2 NA     <NULL>       NA     <NULL>       TRTA     Placebo            N        
#>  3 NA     <NULL>       NA     <NULL>       TRTA     Placebo            p        
#>  4 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… n        
#>  5 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… N        
#>  6 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline High D… p        
#>  7 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… n        
#>  8 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… N        
#>  9 NA     <NULL>       NA     <NULL>       TRTA     Xanomeline Low Do… p        
#> 10 TRTA   Placebo      NA     <NULL>       AESOC    GENERAL DISORDERS… n        
#> # ℹ 2,384 more rows
#> # ℹ 6 more variables: context <chr>, stat_label <chr>, stat <list>,
#> #   fmt_fun <list>, warning <list>, error <list>

Arguments

Value

Note

See also

Examples