I have a proc logistic model with an interaction term between a binary variable and a continuous variable. I'm successfully using the ODDSRATIO statement to have SAS calculate the OR and 95% CIs for the continuous variable at the two different levels of the binary variable.
How can I get SAS to give me (or how can I easily calculate) the p-value associated with these ORs?
While the ODDSRATIO statement in LOGISTIC does not provide p-values, you could fit the same model in PROC GLIMMIX and use the ODDRATIO and SLICEDIFF= options to get the p-values. For details and an example, see this usage note: