Clinical Demographics Table

Demographic summary tables appear in nearly every clinical trial report. This example builds one from scratch with two treatment arms, categorical and continuous variables, missing data, and regulatory-style footnotes.

The data

using StyledTables, DataFrames, SummaryTables

demo = DataFrame(
    variable = [
        "Sex", "Sex", "Sex",
        "Age (years)", "Age (years)", "Age (years)",
        "Race", "Race", "Race", "Race",
    ],
    category = [
        "Male", "Female", "Unknown",
        "Mean (SD)", "Median", "Range",
        "White", "Black or African American", "Asian", "Other/Unknown",
    ],
    placebo_n = [22, 28, missing, "48.3 (12.1)", "47", "24–72", 36, 8, 3, 3],
    treatment_n = [19, 30, 1, "49.7 (11.8)", "50", "22–74", 32, 9, 5, 4],
)

10×4 DataFrame

Row	variable	category	placebo_n	treatment_n
	String	String	Any	Any
1	Sex	Male	22	19
2	Sex	Female	28	30
3	Sex	Unknown	missing	1
4	Age (years)	Mean (SD)	48.3 (12.1)	49.7 (11.8)
5	Age (years)	Median	47	50
6	Age (years)	Range	24–72	22–74
7	Race	White	36	32
8	Race	Black or African American	8	9
9	Race	Asian	3	5
10	Race	Other/Unknown	3	4

Step 1: Basic stub table with multiline column labels

Mark :category as the stub, group rows by :variable, and use Multiline for two-line column headers.

tbl = StyledTable(demo)
tab_stub!(tbl, :category)
tab_rowgroup!(tbl, :variable)
cols_hide!(tbl, :variable)
cols_label!(tbl,
    :placebo_n => Multiline("Placebo (N=50)", "n (%)"),
    :treatment_n => Multiline("Treatment (N=50)", "n (%)"),
)
render(tbl)

	Placebo (N=50) n (%)	Treatment (N=50) n (%)
Sex
Male	22	19
Female	28	30
Unknown	missing	1
Age (years)
Mean (SD)	48.3 (12.1)	49.7 (11.8)
Median	47	50
Range	24–72	22–74
Race
White	36	32
Black or African American	8	9
Asian	3	5
Other/Unknown	3	4

Step 2: Add header, stub label, missing handling, and notes

sub_missing!(tbl, "—")
tab_footnote!(tbl, "Percentages computed on non-missing observations" => [:placebo_n, :treatment_n])
tab_sourcenote!(tbl, "Abbreviations: SD = standard deviation; N = total per arm")
render(tbl)

	Placebo (N=50) n (%)1	Treatment (N=50) n (%)1
Sex
Male	22	19
Female	28	30
Unknown	—	1
Age (years)
Mean (SD)	48.3 (12.1)	49.7 (11.8)
Median	47	50
Range	24–72	22–74
Race
White	36	32
Black or African American	8	9
Asian	3	5
Other/Unknown	3	4
Abbreviations: SD = standard deviation; N = total per arm
1 Percentages computed on non-missing observations

Row groups label the variable category, the stub column names each subgroup, and multiline headers identify each treatment arm and statistic type. The missing Placebo Unknown count appears as an em dash.