Manipulating labels and levels in tbl_regression (gtsummary) with multinom (nnet) - r

Question

It seems the functions for manipulating labels and levels in tbl_regression tables do not work with multinom from the nnet library. Here is reproducable code with the resulting tables. Does anyone have a clue how to get my labels bold and italicized in multinom? Thanks!
library(nnet)
library(gtsummary)
library(tidyverse)
# Create a sample data frame
set.seed(123)
df <- data.frame(
y = sample(c(0, 1, 2), 100, replace = TRUE),
var1 = rnorm(100),
var2 = rnorm(100),
var3 = rnorm(100)
)
# Fit a multinomial logistic regression model with nnet
model <- multinom(y ~ var1 + var2 + var3, data = df)
# Create a summary table with tbl_regression
model_tab <- tbl_regression(model,
exponentiate = TRUE) %>%
bold_labels() %>%
italicize_labels()
model_tab
# Fit a linear regression model
model_2 <- glm(var1 ~ var2 + var3, data = df, family = "gaussian")
# Create a summary table with tbl_regression
model_2_tab <- tbl_regression(model_2) %>%
bold_labels() %>%
italicize_labels()
model_2_tab

Answer 1

If an html document is alright, as I suggested in the comments, you can do it with CSS. In particular, in an RMarkdown document, you would need to put a code chunk that looks like this:
```{css echo=FALSE}
td.gt_row.gt_left{
font-weight: bold;
font-style: italic;
}
```
Then the you could put your code chunk for the table:
```{r echo=FALSE, results="asis"}
tbl_regression(model, exponentiate = TRUE) %>%
bold_labels() %>%
italicize_labels()
```
and in the rendered html document, the labels would show up in bold and italics:
For the categorical variables, the text is indented, so you could differentially change the row headings that are indented. The second block of code in the chunk below sets the indented entries to normal font.
```{css echo=FALSE}
td.gt_row.gt_left{
font-weight: bold;
font-style: italic;
}
td.gt_row.gt_left[style*="text-indent: 10px"]{
font-weight: normal;
font-style: normal;
}
```
I made var1 categorical and here's the resulting table.