The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

Get Started with ggforestplotR

library(ggforestplotR)
library(ggplot2)

ggforestplotR is built for coefficient-driven forest plots that stay inside a normal ggplot2 workflow.

Choose a workflow

Use the package in one of two ways:

  1. Start from a coefficient table and map the required columns directly to the plot.
  2. Start from a fitted model and let tidy_forest_model() or ggforestplot() call broom::tidy() and create the plot.

This article covers some basic examples and the minimum data you need.

Start from a coefficient table

The simplest input is a data frame with a column for terms, estimates, and confidence limits. If your columns use different names, you can map them explicitly.

There is also functionality to rename term labels and sort terms to your liking, among other things.

basic_coefs <- data.frame(
  term = c("Age", "BMI", "Treatment"),
  estimate = c(0.10, -0.08, 0.34),
  conf.low = c(0.02, -0.16, 0.12),
  conf.high = c(0.18, 0.00, 0.56)
)

ggforestplot(basic_coefs,
             term_labels = c("Age" = "age", "BMI" = "bmi", "Treatment" = "trt"),
             sort_terms = "descending")

Add grouped sections and row striping

Use grouping when you want related variables separated into labeled panels. Add striped_rows = TRUE to color alternating rows in the plot.

sectioned_coefs <- data.frame(
  term = c("Age", "BMI", "Smoking", "Stage II", "Stage III", "Nodes"),
  estimate = c(0.10, -0.08, 0.20, 0.34, 0.52, 0.28),
  conf.low = c(0.02, -0.16, 0.05, 0.12, 0.20, 0.06),
  conf.high = c(0.18, 0.00, 0.35, 0.56, 0.84, 0.50),
  section = c("Clinical", "Clinical", "Clinical", "Tumor", "Tumor", "Tumor")
)

ggforestplot(
  sectioned_coefs,
  grouping = "section",
  striped_rows = TRUE,
  stripe_fill = "grey94",
  grouping_strip_position = "right",
  sort_terms = "ascending"
)

Add a summary table

Use add_forest_table() to add a summary table to your forest plot.

tabled_coefs <- data.frame(
  term = c("Age", "BMI", "Smoking", "Stage II", "Stage III"),
  estimate = c(0.12, -0.10, 0.18, 0.30, 0.46),
  conf.low = c(0.03, -0.18, 0.04, 0.10, 0.18),
  conf.high = c(0.21, -0.02, 0.32, 0.50, 0.74),
  sample_size = c(120, 115, 98, 87, 83)
)

ggforestplot(tabled_coefs, striped_rows = TRUE) +
  add_forest_table(
    position = "left",
    column_labels = c("term" = "Variable", "sample_size" = "N", "estimate" = "Beta (95% CI)"),
    columns = c("term", "sample_size", "estimate"),
    estimate_digits = 2,
    interval_digits = 3
  )

Add split summary tables

Use add_split_table() to create a more traditional looking forest plot, with summary data on either side of the plot.

ggforestplot(tabled_coefs, n = "sample_size", striped_rows = T) +
  add_split_table(
    left_columns = c("term", "n"),
    right_columns = c("estimate"),
    column_labels = c("term" = "Variable", "estimate" = "Beta (95% CI)")
  )

Start from a fitted model

If broom is installed, ggforestplot() can work directly from a fitted model.

fit <- lm(mpg ~ wt + hp + qsec, data = mtcars)

ggforestplot(fit, sort_terms = "descending",
             term_labels = c("wt" = "Weight"),
             striped_rows = T) +
  scale_x_continuous(breaks = seq(-6,2,1)) +
  add_forest_table()

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.