plot_reg_combine: Visualize Univariate and Multivariate Regression Side-by-Side

In gtregression: Tools for Creating Publication-Ready Regression Tables

Visualize Univariate and Multivariate Regression Side-by-Side

Description

Generates side-by-side plots to compare univariate & multivariable results

Usage

plot_reg_combine(
  tbl_uni,
  tbl_multi,
  title_uni = NULL,
  title_multi = NULL,
  ref_line = 1,
  order_y = NULL,
  log_x = FALSE,
  point_color = "#1F77B4",
  errorbar_color = "#4C4C4C",
  base_size = 14,
  show_ref = TRUE,
  xlim_uni = NULL,
  breaks_uni = NULL,
  xlim_multi = NULL,
  breaks_multi = NULL
)

Arguments

tbl_uni	A 'gtsummary' object from 'uni_reg()' etc.,
tbl_multi	A 'gtsummary' object from 'multi_reg()'.
title_uni	Optional plot title for the univariate model
title_multi	Optional plot title for the multivariable mode
ref_line	Numeric value for the reference line (default = 1).
order_y	Optional character vector to manually order the y-axis labels.
log_x	Logical. If 'TRUE', x-axis is log-transformed (default = FALSE).
point_color	Optional color for plot points.
errorbar_color	Optional color for error bars.
base_size	Numeric. Base font size for plot text elements.
show_ref	Logical. If 'TRUE', includes reference categories
xlim_uni	Optional numeric vector to set x-axis limits for uni plot.
breaks_uni	Optional numeric vector to set x-axis breaks for uni plot.
xlim_multi	Optional numeric vector to set x-axis limits for multi plot
breaks_multi	Optional numeric vector to set x-axis breaks- multi plot.

Value

A 'ggplot2' object with two forest plots displayed side-by-side.

Examples

if (requireNamespace("mlbench", quietly = TRUE)) {
  data("PimaIndiansDiabetes2", package = "mlbench")
  library(dplyr)
  library(gtregression)

  # Prepare data
  pima <- PimaIndiansDiabetes2 |>
    mutate(
      diabetes = ifelse(diabetes == "pos", 1, 0),
      bmi_cat = cut(
        mass,
        breaks = c(-Inf, 18.5, 24.9, 29.9, Inf),
        labels = c("Underweight", "Normal", "Overweight", "Obese")
      ),
      age_cat = cut(
        age,
        breaks = c(-Inf, 29, 49, Inf),
        labels = c("Young", "Middle-aged", "Older")
      )
    )

  # Univariate logistic regression
  uni_rr <- uni_reg(
    data = pima,
    outcome = "diabetes",
    exposures = c("age_cat", "bmi_cat"),
    approach = "logit"
  )

  # Multivariable logistic regression
  multi_rr <- multi_reg(
    data = pima,
    outcome = "diabetes",
    exposures = c("age_cat", "bmi_cat"),
    approach = "logit"
  )

  # Combined plot
  plot_reg_combine(uni_rr, multi_rr)
}

gtregression documentation built on Aug. 18, 2025, 5:23 p.m.