Skip to contents

One-Click Baseline Characteristics Table (gtsummary version)

Source: R/08_baseline.R
make_baseline_table.Rd

Automatically generates baseline characteristics table with counts (percentages) for categorical variables and appropriate statistics for continuous variables based on normality testing. Outputs a three-line table in Word format with P-values and Overall column.

Usage

make_baseline_table(
  data,
  outcome,
  vars = NULL,
  file,
  pct_digits = 2,
  cont_digits = 2,
  p_digits = 3,
  label_list = NULL,
  normality_test_method = "shapiro",
  normality_alpha = 0.05,
  export_normality = TRUE,
  normality_file = NULL
)

Arguments

data

Data frame containing grouping variable and all variables to summarize

outcome

Character scalar, grouping variable name (typically binary or factor)

vars

Character vector, variables to display; if NULL (default) displays all variables except outcome

file

Character, output Word file name (with path); parent directory must exist

pct_digits

Decimal places for percentages, default 2

cont_digits

Decimal places for continuous variables, default 2

p_digits

Decimal places for P-values, default 3

label_list

Named list, variable labels, e.g., list(age = "Age (years)"); if NULL, uses original column names

normality_test_method

Character, method for normality test ("shapiro" or "ks"), default "shapiro"

normality_alpha

Numeric, significance level for normality test, default 0.05

export_normality

Logical, whether to export normality test results to a separate CSV file, default TRUE

normality_file

Character, output CSV file name for normality test results; if NULL, uses same base name as main file

Value

gtsummary object (invisible), convenient for subsequent piping operations

Details

  • Categorical variables displayed as n(%)

  • Continuous variables: if all groups are normal, display as mean (SD); otherwise display as median (Q1, Q3)

  • Missing values not displayed by default; P-values use appropriate tests (chi-square/t/ANOVA/Kruskal-Wallis)

  • Final table exported via flextable, ready for use in papers or appendices

  • Normality test results can be exported to CSV for reference

Examples

if (FALSE) { # \dontrun{
make_baseline_table(
  data     = mydata,
  outcome  = "Group",
  vars     = c("age", "sex", "smoke", "sbp"),
  file     = "Table1_Baseline.docx",
  label_list = list(age = "Age (years)", sbp = "Systolic BP (mmHg)")
)
} # }