Descriptive Tables

Start with a table that people can actually read. descriptive_table() creates publication-ready summaries for continuous, categorical, and mixed exposure sets.

library(gtregression)
library(dplyr)

data("data_birthwt", package = "gtregression")

birthwt_data <- data_birthwt |>
  mutate(
    race = factor(race, levels = c(1, 2, 3),
                  labels = c("White", "Black", "Other")),
    smoke = factor(smoke, levels = c(0, 1), labels = c("No", "Yes")),
    ht = factor(ht, levels = c(0, 1), labels = c("No", "Yes")),
    ui = factor(ui, levels = c(0, 1), labels = c("No", "Yes")),
    low = factor(low, levels = c(0, 1), labels = c("Normal BW", "Low BW")),
    ptl_cat = factor(ifelse(ptl > 0, "Yes", "No"), levels = c("No", "Yes")),
    ftv_cat = factor(case_when(
      ftv == 0 ~ "None",
      ftv == 1 ~ "One",
      ftv >= 2 ~ "Two or more"
    ), levels = c("None", "One", "Two or more"))
  )

birthwt_exposures <- c(
  "age", "lwt", "race", "smoke", "ht", "ui", "ptl_cat", "ftv_cat"
)

attr(birthwt_data$age, "label") <- "Maternal age"
attr(birthwt_data$lwt, "label") <- "Maternal weight"
attr(birthwt_data$race, "label") <- "Maternal race"
attr(birthwt_data$smoke, "label") <- "Smoking during pregnancy"
attr(birthwt_data$ht, "label") <- "Hypertension"
attr(birthwt_data$ui, "label") <- "Uterine irritability"
attr(birthwt_data$ptl_cat, "label") <- "Previous preterm labour"
attr(birthwt_data$ftv_cat, "label") <- "First trimester visits"

Column Percentages

Use column percentages when the table is grouped by outcome or another column. Continuous variables are shown as median (IQR) by default, and variable labels are picked up automatically when variables have a "label" attribute.

desc_column <- descriptive_table(
  data = birthwt_data,
  exposures = birthwt_exposures,
  by = "low",
  percent = "column",
  show_overall = "last",
  theme = clinical
)

desc_column
## a flextable object.
## col_keys: `Characteristic`, `Normal BW, N=130`, `Low BW, N=59`, `Overall, N=189` 
## header has 1 row(s) 
## body has 22 row(s) 
## original dataset sample: 
## 'data.frame':    22 obs. of  4 variables:
##  $ Characteristic  : chr  "Maternal age" "Maternal weight" "Maternal race" "  White" ...
##  $ Normal BW, N=130: chr  "23.0 (19.0-28.0)" "123.5 (113.0-147.0)" "" "73 (56.2%)" ...
##  $ Low BW, N=59    : chr  "22.0 (19.5-25.0)" "120.0 (104.0-130.0)" "" "23 (39.0%)" ...
##  $ Overall, N=189  : chr  "23.0 (19.0-26.0)" "121.0 (110.0-140.0)" "" "96 (50.8%)" ...

Row Percentages

Use row percentages when the question is how each exposure level is distributed across groups. Common option values can be written with or without quotes.

descriptive_table(
  data = birthwt_data,
  exposures = birthwt_exposures,
  by = "low",
  percent = "row",
  show_overall = "first",
  show_missing = no,
  theme = striped
)$table

Characteristic

Overall, N=189

Normal BW, N=130

Low BW, N=59

Maternal age

23.0 (19.0-26.0)

23.0 (19.0-28.0)

22.0 (19.5-25.0)

Maternal weight

121.0 (110.0-140.0)

123.5 (113.0-147.0)

120.0 (104.0-130.0)

Maternal race

White

96

73 (76.0%)

23 (24.0%)

Black

26

15 (57.7%)

11 (42.3%)

Other

67

42 (62.7%)

25 (37.3%)

Smoking during pregnancy

No

115

86 (74.8%)

29 (25.2%)

Yes

74

44 (59.5%)

30 (40.5%)

Hypertension

No

177

125 (70.6%)

52 (29.4%)

Yes

12

5 (41.7%)

7 (58.3%)

Uterine irritability

No

161

116 (72.0%)

45 (28.0%)

Yes

28

14 (50.0%)

14 (50.0%)

Previous preterm labour

No

159

118 (74.2%)

41 (25.8%)

Yes

30

12 (40.0%)

18 (60.0%)

First trimester visits

None

100

64 (64.0%)

36 (36.0%)

One

47

36 (76.6%)

11 (23.4%)

Two or more

42

30 (71.4%)

12 (28.6%)

Categorical variables shown as n (%); percentages are by row (Overall shows counts).

Continuous variables shown as Median (IQR).

Summary Choices

Use statistic when continuous variables need a different summary. A single value applies to all numeric variables. A named vector lets you mix summaries, including treating numeric ordinal variables as categorical.

descriptive_table(
  data = birthwt_data,
  exposures = c("age", "lwt", "ftv", "smoke"),
  by = low,
  statistic = c(
    age = mean,
    lwt = median,
    ftv = categorical
  ),
  percent = column,
  show_missing = no
)
## a flextable object.
## col_keys: `Characteristic`, `Normal BW, N=130`, `Low BW, N=59` 
## header has 1 row(s) 
## body has 12 row(s) 
## original dataset sample: 
## 'data.frame':    12 obs. of  3 variables:
##  $ Characteristic  : chr  "Maternal age" "Maternal weight" "ftv" "  0" ...
##  $ Normal BW, N=130: chr  "23.7 (5.6)" "123.5 (113.0-147.0)" "" "64 (49.2%)" ...
##  $ Low BW, N=59    : chr  "22.3 (4.5)" "120.0 (104.0-130.0)" "" "36 (61.0%)" ...

The quoted form is equivalent and often clearer in saved scripts:

descriptive_table(
  data = birthwt_data,
  exposures = c("age", "lwt", "ftv", "smoke"),
  by = "low",
  statistic = c(
    age = "mean",
    lwt = "median",
    ftv = "categorical"
  )
)

Output Format

flextable is the default because it behaves well in Word workflows. Use format = gt when the output is mainly for HTML or pkgdown.

desc_gt <- descriptive_table(
  data = birthwt_data,
  exposures = birthwt_exposures,
  by = "low",
  percent = "column",
  format = gt
)

desc_gt$table
Characteristic Normal BW, N=130 Low BW, N=59
Maternal age 23.0 (19.0-28.0) 22.0 (19.5-25.0)
Maternal weight 123.5 (113.0-147.0) 120.0 (104.0-130.0)
Maternal race
White 73 (56.2%) 23 (39.0%)
Black 15 (11.5%) 11 (18.6%)
Other 42 (32.3%) 25 (42.4%)
Smoking during pregnancy
No 86 (66.2%) 29 (49.2%)
Yes 44 (33.8%) 30 (50.8%)
Hypertension
No 125 (96.2%) 52 (88.1%)
Yes 5 (3.8%) 7 (11.9%)
Uterine irritability
No 116 (89.2%) 45 (76.3%)
Yes 14 (10.8%) 14 (23.7%)
Previous preterm labour
No 118 (90.8%) 41 (69.5%)
Yes 12 (9.2%) 18 (30.5%)
First trimester visits
None 64 (49.2%) 36 (61.0%)
One 36 (27.7%) 11 (18.6%)
Two or more 30 (23.1%) 12 (20.3%)
Categorical variables shown as n (%); percentages are by column.
Continuous variables shown as Median (IQR).

What To Inspect