I am having an issue with faceting a dataset looking at different medical devices over a certain time point. The figure I am looking to do uses a facet_wrap with two variables (I know I could also use facet_grid but prefer the appearance of facet_wrap in this case)
The issue I have is:
1.) The orientation of the strip labels, where I would like the top strip to be below the other one and
2) that I would like the bottom strip label to cross both columns as it is the same.
An example dataset and code is below:
library(tidyverse)
library(ggplot2)
dt <- tibble(device_type = c("Stent","Stent","Stent","Stent", "Pace","Pace","Pace","Pace"),
days = c(5,10,15,2,4,6,6,8),
generation = c("First", "Second","First", "Second","First", "Second","First", "Second"),
year = c("Before", "Before","After", "After","Before", "Before","After", "After"))
dt %>%
ggplot(aes(x=generation, y= days))+
geom_bar(stat = "identity")+
facet_wrap(~device_type + year, strip.position = "bottom", nrow = 1)+
theme(strip.placement = "outside")
which gives the plot: 
What I am trying to get the orientation of the facet strip labels to look like is here:
I first tried changing the position of the variables in the facet_wrap line to facet_wrap(~year + device_type, strip.position = "bottom", nrow = 1) but this alters the plot in such a way that it splits the grouping into years which is not ideal for the actual dataset im using this on.
I found this post Combine multiple facet strips across columns in ggplot2 facet_wrap utilizing facet_nested but have been unable to get this to work with the switching of the strip labels. Any help or ideas would be greatly appreciated.
>Solution :
This could be easily done with ggh4x package written by teunbrand:
Using facet_nested_wrap function:
You can change which one you want to combine, just change the order in facet_nested_wrap:
library(tidyverse)
#install.packages("ggh4x")
library(ggh4x)
dt %>%
ggplot(aes(x=generation, y= days))+
geom_bar(stat = "identity")+
facet_nested_wrap(~year + device_type, nrow = 1, ncol=4)
