gtregression

Publication-ready regression and survival analysis tables, plots, and forest plots for real-world health data. Fit models, compare estimates, visualise results, and export manuscript-ready outputs without hand-formatting every coefficient.

Start the workflow Explore functions

Publication-Ready Regression, Survival, and Mediation Outputs

gtregression helps you move from model to manuscript: fit regression models, produce clean tables, visualise estimates, merge outputs, and export results without hand-formatting every coefficient.

It supports logistic, log-binomial, Poisson, robust Poisson, negative binomial, linear, Cox, parametric survival, and causal mediation workflows, including adjusted and stratified models.

Build	What you get
Descriptive tables	Grouped summaries with row or column percentages
Regression tables	Crude, adjusted, stratified, linear, Cox, and parametric survival outputs
Survival analysis	Kaplan-Meier curves, survival summaries, RMST, log-rank tests, Cox PH checks, and survival predictions
Mediation analysis	Direct, indirect, total, and proportion mediated effects with causal caveats
Visualisations	Regression plots, survival curves, fitted survival curves, and forest tables
Interpretation helpers	Confounding, interaction, mediation, convergence, collinearity, model selection, and survival diagnostics
Exports	HTML, PDF, PNG, and Word-ready outputs

From Data to Manuscript

One connected workflow

Start with the question. Finish with a result you can use.

Each step leaves an inspectable object behind, so beginners have a clear path and experienced analysts retain full control.

Prepare

Check variables, labels, levels, and missing data.

dissect(data) Analysis-ready data

Describe

Build a clear baseline table before modelling.

descriptive_table(...) Table 1

Model

Fit crude, adjusted, stratified, or survival models.

uni_reg() + multi_reg() Effect estimates

Interpret

Review assumptions, confounding, interaction, and fit.

check_*() + compare_models() Defensible model

Publish

Merge, visualise, and export polished outputs.

forest_reg() + save_table() Manuscript-ready output

Why It Exists

Many students, researchers, and public health analysts need regression outputs that are readable, reproducible, and report-ready. gtregression keeps the R syntax approachable while preserving transparent model objects underneath.

Built on Trusted R Packages

gtregression is intentionally a readable interface over established R packages. The package uses widely trusted modelling, tidying, plotting, and reporting tools so users can inspect fitted models and understand the statistical engines behind each output.

Area	Core packages used
Data handling and tidy workflows	`dplyr`, `purrr`, `tibble`, `rlang`
Model fitting	`stats`, `MASS`, `survival`, `risks`, `logistf`
Robust and diagnostic inference	`sandwich`, `lmtest`, `broom`, `broom.helpers`
Tables and Word-ready reporting	`flextable`, `officer`, `gt`
Figures and forest plots	`ggplot2`, `patchwork`, `forestploter`, `scales`
Optional development and checking tools	`testthat`, `knitr`, `rmarkdown`, `pkgdown`, `car`, `forcats`, `ggtext`

The user-facing functions return objects with fitted models, table bodies, and display metadata that advanced users can audit, modify, or reuse.

Install

# CRAN
install.packages("gtregression")

# Development version
remotes::install_github("ThinkDenominator/gtregression")

Five-Minute Workflow

library(gtregression)
library(dplyr)

data("data_birthwt", package = "gtregression")

birthwt_data <- data_birthwt |>
  mutate(
    race = factor(race, levels = c(1, 2, 3),
                  labels = c("White", "Black", "Other")),
    smoke = factor(smoke, levels = c(0, 1), labels = c("No", "Yes")),
    ht = factor(ht, levels = c(0, 1), labels = c("No", "Yes")),
    ui = factor(ui, levels = c(0, 1), labels = c("No", "Yes")),
    low = factor(low, levels = c(0, 1), labels = c("Normal BW", "Low BW"))
  )

exposures <- c("age", "lwt", "race", "smoke", "ht", "ui")

attr(birthwt_data$age, "label") <- "Maternal age"
attr(birthwt_data$lwt, "label") <- "Maternal weight"
attr(birthwt_data$smoke, "label") <- "Smoking during pregnancy"

desc <- descriptive_table(
  birthwt_data,
  exposures = exposures,
  by = "low",
  percent = "column",
  show_overall = "last"
)

uni <- uni_reg(
  birthwt_data,
  outcome = "low",
  exposures = exposures,
  approach = "logit"
)

multi <- multi_reg(
  birthwt_data,
  outcome = "low",
  exposures = c("smoke", "ht", "ui"),
  adjust_for = c("age", "lwt", "race"),
  approach = "logit"
)

plot_reg(multi, title = "Adjusted Regression for Low Birth Weight")

forest_reg = forest_reg(forest_df(uni, multi)) 

merge_tables(desc, uni, multi)

Variable labels set with attr(x, "label") or labelled::var_label() are used automatically in display tables and plots, while original column names remain available internally for merging, modification, and testing.

Objects stay inspectable:

desc$table
uni$table
multi$table
multi$models

Optional model-fit statistics can be requested without changing the publication table:

uni_stats <- uni_reg(
  data = birthwt_data,
  outcome = "low",
  exposures = exposures,
  approach = "logit",
  model_stats = TRUE
)

uni_stats$model_stats

Browse by Task

Task	Start here
First workflow	Start Here
Descriptive summaries	Descriptive Tables
Regression tables	Regression Tables
Survival analysis	Survival Analysis
Causal mediation	Causal Mediation
Visualise estimates	Visualise Results
Stratified models	Stratified Analysis
Diagnostics and selection	Diagnostics
Confounding and interaction	Confounding & Interaction
Merge and export	Customize and Export

Function Map

Workflow	Functions
Describe	`descriptive_table()`, `dissect()`
Model	`uni_reg()`, `multi_reg()`, `cox_reg()`, `surv_reg()`
Survival	`km_plot()`, `km_risk_table()`, `survival_summary()`, `survival_quantiles()`, `survival_prob()`, `rmst_table()`, `logrank_test()`, `check_ph()`, `surv_model_compare()`, `plot_surv_fit()`, `surv_predict()`
Stratify	`stratified_uni_reg()`, `stratified_multi_reg()`
Visualise	`plot_reg()`, `plot_reg_combine()`, `forest_df()`, `forest_reg()`
Diagnose	`check_convergence()`, `check_collinearity()`, `check_ph()`, `select_models()`
Interpret	`identify_confounder()`, `interaction_models()`, `mediation_analysis()`, `plot_mediation()`
Polish and export	`modify_table()`, `merge_tables()`, `save_table()`, `save_plot()`, `save_docx()`

Citation

If you use gtregression in your work, please cite it as:

Polani R, Eliyas SK, Sakthivel M, Kaviprawin M, Krishnamoorthy Y, Majella MG. gtregression: Tools for Creating Publication-Ready Regression Tables. Zenodo. https://doi.org/10.5281/zenodo.16905350

Acknowledgements

gtregression builds on the R ecosystem, especially stats, survival, MASS, risks, logistf, broom, broom.helpers, sandwich, lmtest, dplyr, purrr, tibble, rlang, flextable, officer, gt, ggplot2, patchwork, forestploter, and scales.