Non-Inferiority Test in Multple Logistic Model


Less is more. Stay pure. Stay poor.
Does any body have any insight or experience running a non-inferiority test based on multiple linear or logistic regression. So perhaps mean and risk differences or odds ratio. I have binary and continuous outcomes, but more interested in binary. Seems like this should be a contrast statement thing.

More specifically I would be interested in seeing an example in R or SAS.