Skip to contents

Design A Descriptive Table

Source: R/table1_design.R
table1_design.Rd

This function generates a design table from which rifttable can generate a descriptive table.

Usage

table1_design(
  data,
  ...,
  by = NULL,
  total = TRUE,
  empty_levels = FALSE,
  na_always = FALSE,
  na_label = "Unknown",
  continuous_type = "median (iqr)",
  binary_type = "outcomes (risk)"
)

Arguments

data

Data set

...

Optional: Variables to include or exclude (using -variable)

by

Optional: Stratification variable. Typically the exposure.

total

Optional: Whether to add the total count at the beginning. Defaults to TRUE.

empty_levels

Optional: Whether to include empty levels of factor variables. Defaults to FALSE.

na_always

Optional: Whether to add the count of missing values for each variable, even if there are none. Defaults to FALSE, i.e., the count of missing values will only be shown if there are any.

na_label

Label for count of missing values. Defaults to "Unknown".

continuous_type

Estimator (type in rifttable design) for continuous variables. Defaults to "median (iqr)".

binary_type

Estimator (type in rifttable design) for binary variables and strata of categorical variables. Defaults to "outcomes (risk)" (count and column proportion).

Value

design tibble that can be passed on to rifttable. Contains an attribute rt_data

so that the dataset does not have to be provided to rifttable another time.

Examples

# Data preparation
cars <- tibble::as_tibble(mtcars) %>%
  dplyr::mutate(
    gear = factor(
      gear,
      levels = 3:5,
      labels = c("Three", "Four", "Five")),
    hp_categorical = dplyr::if_else(
      hp > 200,
      true = ">200 hp",
      false = "<=200 hp"))
# Label some variables. Better alternative: labelled::set_variable_labels()
attr(cars$hp, "label") <- "Horsepower"
attr(cars$hp_categorical, "label") <- "Horsepower"
attr(cars$am, "label") <- "Automatic transmission"
attr(cars$gear, "label") <- "Gears"

# Generate table "design"
design <- cars %>%
  table1_design(
    hp, hp_categorical, mpg, am,
    by = gear)

# Use "design" to create a descriptive table.
design %>%
  rifttable(diff_digits = 0)
#> # A tibble: 7 × 4
#>   Gears                    Three            Four           Five            
#>   <chr>                    <chr>            <chr>          <chr>           
#> 1 "N"                      "15"             "12"           "5"             
#> 2 "Horsepower"             "180 (150, 210)" "94 (66, 110)" "175 (113, 264)"
#> 3 "Horsepower"             ""               ""             ""              
#> 4 "  <=200 hp"             "10 (67%)"       "12 (100%)"    "3 (60%)"       
#> 5 "  >200 hp"              "5 (33%)"        "0 (0%)"       "2 (40%)"       
#> 6 "mpg"                    "16 (14, 18)"    "23 (21, 28)"  "20 (16, 26)"   
#> 7 "Automatic transmission" "0 (0%)"         "8 (67%)"      "5 (100%)"