Reputation: 1
I would like to use tbl_regression in gtsummary to exponentiate for my ORs, but at different unit values. For example, I have a logistic regression model in which for some predictors I would like to express the OR per 5 or 10 unit increase (not just 1 unit increase). Is there anyway to do so?
Per my understanding of tbl_regression, I do not see a way to customize the exponentiate function.
Upvotes: 0
Views: 280
Reputation: 11754
It is possible from within gtsummary, but it gets more complicated with exponentiation. I recommend just handling this before the model is passed to tbl_regression()
. See example
library(gtsummary)
packageVersion("gtsummary")
#> [1] '1.7.0'
glm(response ~ identity(age/10) + grade, data = trial, family = binomial) |>
tbl_regression(
exponentiate = TRUE,
label = list("identity(age/10)" = "Age per 10yrs")
) |>
as_kable() # convert to kable to display on StackOverflow
Characteristic | OR | 95% CI | p-value |
---|---|---|---|
Age per 10yrs | 1.21 | 0.97, 1.52 | 0.10 |
Grade | |||
I | — | — | |
II | 0.85 | 0.39, 1.85 | 0.7 |
III | 1.01 | 0.47, 2.16 | >0.9 |
Created on 2023-03-01 with reprex v2.0.2
Upvotes: 0