# Set the working directory to the location of the stroke paper files
setwd("C:/Users/Administrator/Desktop/Stroke_Paper")
data <- read.csv("mytable.csv")

#-----------------------Data Format Transformation-------------------------
str(data)

# Convert columns 5 to 23 to factor type, excluding the outcome variable
data[,c(5:23)] <- lapply(data[,c(5:23)], as.factor)
str(data)
summary(data)
# Note: The outcome variable should not be converted to factor type

# Display categorical data
table(data$zaifa)  # Categorical data

#-------------------------1. Baseline Information Table---------------------------
# Continuous variables are reported as [Normal] mean ± standard deviation (SD) or [Non-normal] median interquartile range (IQR)
# Categorical variables are represented as count (percentage)

## ----Normality Test----

# Shapiro-Wilk test (sample size < 5000, P < 0.05 indicates non-normality)
lapply(data[,c(24:76)], shapiro.test)

## ----Create Baseline Information Table----
library(tableone)
tab1 <- CreateTableOne(
  vars = colnames(data[,c(5:76)]),  # All variables to be described
  factorVars = colnames(data[,c(5:23)]),  # Categorical variables
  strata = "zaifa",  # Stratification variable
  data = data,
  addOverall = TRUE
)
tab1

# Print the table, handling non-normal variables and formatting
tab <- print(tab1,
             nonnormal = colnames(data[,c(24:76)]),  # Variables with non-normal distribution
             noSpaces = TRUE,  # Remove spaces added for alignment
             showAllLevels = TRUE)  # Show all levels of categorical variables

# Write the table to a CSV file in English
write.csv(tab, "C:/Users/Administrator/Desktop/Stroke_Paper/Baseline_Information_Table.csv")