flexTab1

Overview

CAVE: This function is still in development mode. This function creates a highly flexible Table 1 for descriptive statistics with options to customize both the statistical options and layout. It supports:

• Statistical Flexibility: Various summary measures can be selected for categorical (absolute, relative) and numerical variables (mean, median, sd, min, max, quartile1, quartile3), arranged in any combination. Missing values frequencies for each variable can be optionally displayed. For group comparisons, options include displaying p-values and standardized mean differences (SMD).

• Custom Layout:The following layout options are available: row sorting, the option to display summary measure units (e.g., n (%), median (SD)), and an empty line after each variable for improved readability. The output can be generated as either a publication-ready "flextable()" object or as a "data.frame", allowing for further modifications or integration with other table formatting packages.

• Group Analysis: Supports single and grouped summaries for comparing treatment arms or other groups. Additionally, a nested group structure is supported, allowing for the comparison of superior groups (A, B, and C), each with two subgroups (x and y).

This package is still an experimental version.

Installation

You can install the development version of flexTab1 from GitHub

# install.packages("devtools")
devtools::install_github("KGutmair/flexTab1")

Usage

Table 1 for one group/treatment arm

library(dplyr)
library(flexTab1)
library(flextable)



# Load pbc data from the survival package
pbc <- survival::pbc

baseline_var <- c("age", "chol", "sex", "stage", "platelet")

pbc <- pbc %>%
     mutate_at(c("stage", "trt", "edema", "hepato"), function(x) as.factor(x)) %>%
     filter(!is.na(trt)) %>%
     mutate(trt = ifelse(trt == "1", "D-penicillmain", "Placebo"))

 tab1_example <- Table1_flex(
     data = pbc,
     variables = baseline_var 
   )

tab1_example %>%
   line_spacing(space = 1.7, part = "all") %>%
   autofit() 

Table 1 for two treatment arms

tab1_example <- Table1_flex(
  data = pbc,
  variables = baseline_var,
  group_var = "trt",
  add_measure_ident = TRUE,
  display_pvalue = TRUE, 
  sort_rows = c("age", "sex", "stage", "chol", "platelet"),
  measures_cat = c("absolute", "relative"),
  measures_num = c("median", "min", "max")
)

tab1_example %>%
   line_spacing(space = 1.7, part = "all") %>%
   autofit() 

Table 1 for nested groups

The Table1_flex also supports a nested group structure, meaning having subgroups within superior groups

baseline_var <- c("age", "chol", "platelet", "stage")
tab1_example <- Table1_flex(
  data = pbc,
  variables = baseline_var,
  group_var = "sex",
  treatment_arm = "trt",
  add_measure_ident = TRUE,
  measures_cat = c("relative"),
  measures_num = c("mean","sd"),
  flextable_output = TRUE
)

tab1_example %>%
   line_spacing(space = 1.7, part = "all") %>%
   autofit()