Append a rowwise summary indicator column — append_rowwise_summary

Low level function that appends a (binary) indicator column to a dataframe, summarising rowwise whether any values in a selection of existing columns match one or more specified values.

Usage

append_rowwise_summary_indicator(
  df,
  cols,
  yes_values,
  new_indicator_colname,
  new_indicator_var_label,
  new_indicator_yes_value = "Yes",
  new_indicator_no_value = "No"
)

Arguments

df: Data frame
cols: Character vector of columns to summarise
yes_values: Vector. new_indicator_colname will be "Yes" if any values in cols (rowwise) match one of these values.
new_indicator_colname: Name of new indicator column.
new_indicator_var_label: Character string to set as variable label for new indicator column.
new_indicator_yes_value: Default is "Yes".
new_indicator_no_value: Default is "No"

Value

A data frame with an additional column, named as per new_indicator_colname.

Examples

# example input df
df <- tibble::tribble(
~med_1, ~med_2, ~med_3, ~IGNORED_COL,
"Beta blocker", NA, NA, "Beta blocker",
"ACE inhibitor", "Beta blocker", NA, "Beta blocker",
"Aspirin", "Aspirin", NA, "Beta blocker",
"Aspirin", "ACE inhibitor", NA, "Beta blocker"
)

# add indicator col
df <- append_rowwise_summary_indicator(df = df,
cols = c("med_1", "med_2", "med_3"),
yes_values = c("Beta blocker", "ACE inhibitor"),
new_indicator_colname = "on_anti_hypertensive",
new_indicator_var_label = "Taking anti-hypertensive medication",
new_indicator_yes_value = "Yes",
new_indicator_no_value = "No")

# print result
print(df)
#> # A tibble: 4 × 5
#>   med_1         med_2         med_3 IGNORED_COL  on_anti_hypertensive
#>   <chr>         <chr>         <lgl> <chr>        <chr>               
#> 1 Beta blocker  NA            NA    Beta blocker Yes                 
#> 2 ACE inhibitor Beta blocker  NA    Beta blocker Yes                 
#> 3 Aspirin       Aspirin       NA    Beta blocker No                  
#> 4 Aspirin       ACE inhibitor NA    Beta blocker Yes                 

# print variable label for new column
print(attributes(df$on_anti_hypertensive)$label)
#> [1] "Taking anti-hypertensive medication"