Chapter 6 Baseline Table

A Table 1 summarising baseline characteristics stratified by case/control status is standard in epidemiological papers. UKBAnalytica wraps the tableone package for this purpose.

6.1 Basic usage

library(UKBAnalytica)
library(data.table)

# Assume `analysis_dt` was produced by build_survival_dataset()
table1 <- create_baseline_table(
  data = analysis_dt,
  case_col = "outcome_status",
  factor_cols = c(
    "sex", "smoking", "drinking",
    "ethnicity", "education"
  ),
  continuous_cols = c(
    "age", "bmi", "sbp", "dbp",
    "ldl", "hdl", "hba1c"
  ),
  test = TRUE
)

print(table1, smd = TRUE)

6.2 Exporting to CSV

The tableone print method returns a character matrix that can be written directly to CSV:

tab_mat <- print(
  table1,
  quote  = TRUE,
  noSpaces = TRUE
)
write.csv(tab_mat, file = "baseline_table1.csv")

6.3 Using with the survival package example data

If you do not have UKB data at hand, you can test the function with the pbc dataset bundled in the survival package:

data(pbc, package = "survival")

table1 <- create_baseline_table(
  data       = pbc,
  case_col   = "trt",
  factor_cols = c("status", "edema", "stage"),
  continuous_cols = c("age", "bili", "albumin"),
  test       = TRUE
)

print(table1)

6.4 Parameters

Argument	Description
data	A data.frame or data.table with the cohort
case_col	Column name for the stratification variable
factor_cols	Variables to treat as categorical
continuous_cols	Variables to treat as continuous
test	Whether to run statistical tests (default FALSE)