This article is about the look — titles, footnotes, the running page header / footer, and cell styling. It assumes the table’s shape is already built (see Structure) and never explains data prep or pagination for its own sake.
Titles and footnotes
Multi-line titles and static footnotes are arguments to tabular():
tabular(cdisc_saf_demo, titles =c("Table 14-2.01","Demographic and Baseline Characteristics","ITT Population"), footnotes =c("Percentages are based on the number of ITT subjects per arm.","BMI = body mass index."))|>cols( variable =col_spec( usage ="group", group_display ="header_row", label =""), stat_label =col_spec(label =""))|>cols_apply(arms, col_spec(align ="decimal"))
Table 14-2.01
Demographic and Baseline Characteristics
ITT Population
placebo
drug_50
drug_100
Total
Age (years)
n
86
96
72
254
Mean (SD)
75.2 (8.59)
76.0 (8.11)
73.8 (7.94)
75.1 (8.25)
Median
76.0
78.0
75.5
77.0
Q1, Q3
69.2, 81.8
71.0, 82.0
70.5, 79.0
70.0, 81.0
Min, Max
52 , 89
51 , 88
56 , 88
51 , 89
Sex, n (%)
F
53 (61.6)
55 (57.3)
35 (48.6)
143 (56.3)
M
33 (38.4)
41 (42.7)
37 (51.4)
111 (43.7)
Race, n (%)
WHITE
78 (90.7)
90 (93.8)
62 (86.1)
230 (90.6)
BLACK OR AFRICAN AMERICAN
8 ( 9.3)
6 ( 6.2)
9 (12.5)
23 ( 9.1)
ASIAN
0
0
0
0
AMERICAN INDIAN OR ALASKA NATIVE
0
0
1 ( 1.4)
1 ( 0.4)
Percentages are based on the number of ITT subjects per arm.
BMI = body mass index.
For an anchored footnote (an auto-numbered superscript on a specific cell or header) use footnote() with a cells_*() location:
base|>footnote("Excludes one subject withdrawn before dosing.", .at =cells_headers(j ="Total"))
placebo
drug_50
drug_100
Totala
Age (years)
n
86
96
72
254
Mean (SD)
75.2 (8.59)
76.0 (8.11)
73.8 (7.94)
75.1 (8.25)
Median
76.0
78.0
75.5
77.0
Q1, Q3
69.2, 81.8
71.0, 82.0
70.5, 79.0
70.0, 81.0
Min, Max
52 , 89
51 , 88
56 , 88
51 , 89
Sex, n (%)
F
53 (61.6)
55 (57.3)
35 (48.6)
143 (56.3)
M
33 (38.4)
41 (42.7)
37 (51.4)
111 (43.7)
Race, n (%)
WHITE
78 (90.7)
90 (93.8)
62 (86.1)
230 (90.6)
BLACK OR AFRICAN AMERICAN
8 ( 9.3)
6 ( 6.2)
9 (12.5)
23 ( 9.1)
ASIAN
0
0
0
0
AMERICAN INDIAN OR ALASKA NATIVE
0
0
1 ( 1.4)
1 ( 0.4)
a Excludes one subject withdrawn before dosing.
The running page header / footer
Regulatory TFLs carry a header on every page (protocol, page X of Y, data-cut). That is preset page chrome, not a title. pagehead / pagefoot each take left / right (and center) vectors; the tokens {page}, {npages}, {program}, {datetime} resolve at render time:
chrome<-base|>preset( pagehead =list( left =c("Analysis Set: Safety", "Protocol: XYZ-123"), right =c("Data cut: 2026-01-15", "Page {page} of {npages}")))chrome
Protocol: XYZ-123 Analysis Set: Safety
Page 1 of 1 Data cut: 2026-01-15
placebo
drug_50
drug_100
Total
Age (years)
n
86
96
72
254
Mean (SD)
75.2 (8.59)
76.0 (8.11)
73.8 (7.94)
75.1 (8.25)
Median
76.0
78.0
75.5
77.0
Q1, Q3
69.2, 81.8
71.0, 82.0
70.5, 79.0
70.0, 81.0
Min, Max
52 , 89
51 , 88
56 , 88
51 , 89
Sex, n (%)
F
53 (61.6)
55 (57.3)
35 (48.6)
143 (56.3)
M
33 (38.4)
41 (42.7)
37 (51.4)
111 (43.7)
Race, n (%)
WHITE
78 (90.7)
90 (93.8)
62 (86.1)
230 (90.6)
BLACK OR AFRICAN AMERICAN
8 ( 9.3)
6 ( 6.2)
9 (12.5)
23 ( 9.1)
ASIAN
0
0
0
0
AMERICAN INDIAN OR ALASKA NATIVE
0
0
1 ( 1.4)
1 ( 0.4)
The running header/footer is page chrome — it repeats on every page of the paged backends (RTF, PDF, DOCX) and does not appear in the single-page HTML preview above. Emit to a paged backend to see it:
emit(chrome, "demographics.pdf")# protocol + page x of y on every page
Stacking direction. Each vector stacks outward from the table: index 1 is the line nearest the table body, later elements move toward the page edge. For a header that reads “Analysis Set” on top and “Protocol” just above the table, put "Protocol…" first. (Easy to invert — check the rendered page.)
Cell styling
style() appends one location + attribute layer. Target a region with a cells_*() constructor; attributes include bold, italic, underline, color, background, and borders via brdr():
base|># the column-header band is bold by default, so style it with colour /# background for a visible change rather than re-applying boldstyle(color ="#1F3B5C", background ="#DBE4F0", .at =cells_headers())|>style(italic =TRUE, .at =cells_group_headers())|># italic section rowsstyle(background ="#F2F2F2", .at =cells_body(j ="Total"))# shade the Total column
placebo
drug_50
drug_100
Total
Age (years)
n
86
96
72
254
Mean (SD)
75.2 (8.59)
76.0 (8.11)
73.8 (7.94)
75.1 (8.25)
Median
76.0
78.0
75.5
77.0
Q1, Q3
69.2, 81.8
71.0, 82.0
70.5, 79.0
70.0, 81.0
Min, Max
52 , 89
51 , 88
56 , 88
51 , 89
Sex, n (%)
F
53 (61.6)
55 (57.3)
35 (48.6)
143 (56.3)
M
33 (38.4)
41 (42.7)
37 (51.4)
111 (43.7)
Race, n (%)
WHITE
78 (90.7)
90 (93.8)
62 (86.1)
230 (90.6)
BLACK OR AFRICAN AMERICAN
8 ( 9.3)
6 ( 6.2)
9 (12.5)
23 ( 9.1)
ASIAN
0
0
0
0
AMERICAN INDIAN OR ALASKA NATIVE
0
0
1 ( 1.4)
1 ( 0.4)
Body filters live on cells_body(): i = (row index), j = (column name), where = (an expression over the data).
Presets: cosmetics and fit
preset() carries the cosmetic defaults — fonts, rules, padding, na_text, paper/orientation/margins — and the width_mode that decides how the table fills the page:
width_mode
Effect
"content"(default)
columns sized to their content
"window"
"auto" columns stretch to fill the printable width (Word “Auto-fit Window”)
Use set_preset() once at the top of a study to make these defaults apply to every table without restating them.
Decimal alignment (col_spec(align = "decimal")) pads numeric cells with non-breaking spaces so the decimal points line up. Padding is measured with the built-in font metrics (preset(decimal_metrics = "afm"), the default), so alignment is exact in Courier and exact to within one padding space in proportional fonts such as Times New Roman or Arial. Markdown output pads by character count instead — the right geometry for a text medium.
