forest plot with grouping need

grouped_forest.R

library(broom)
library(tidyverse)
library(viridis)


fit <- lm(mpg ~ ., data = mtcars)

fit_dt <- tidy(fit, conf.int = T) 

# reverse the order so it looks right on plot
fit_dt$term <- factor(fit_dt$term, rev(fit_dt$term))

# I'm creating some fake grouping info
fit_dt$cat <- c(rep("type a", 4), rep("type b", 4), rep("type c", 3))

ggplot(fit_dt, aes(x = term, y = estimate, ymin = conf.low, ymax = conf.high)) +
  geom_hline(yintercept = 0, color = "gray") + 
  geom_pointrange(aes(color = cat)) +
  coord_flip() +
  theme_classic() +
  scale_color_viridis(end = 0.8, discrete = T)