I am working on a multivariable logistic regression model in R. My goal is to compare Mortality for a female cohort group using males as a reference. I have specified males to be 0 and females to be 1. I am having trouble understanding the output and how to calculate the adjusted odds ratio. For some variables I am receiving an odds ratio of 0 and a really large CI. R does throw the error:
glm.fit: fitted probabilities numerically 0 or 1 occurred If anyone could help me understand how to calculate adjusted odds ratio and how to use the multivariable logistic regression using the males as a reference I would greatly appreciate it.
My formula is as follows:
glm(formula = Outcome ~ SEX + DIABETES + STEROID, family = binomial, data = df) Outcome is a post op variable we are trying to predict using certain pre op variables.