This should be a very simple question, but I cannot seem to figure out how to do it. I ran the crude ORs for each variable to assess the odds that persons would get tested for X disease at one clinic site vs. another (i.e., there is no reference group since I am comparing male who got tested at one clinic to males who got tested at another, females who got tested at one clinic vs. another, etc.). The table we put together looks like: Male: OR = 2.0 (95% CI: 1.5-2.5) Female: OR 2.5 (95% CI: 1.8-.3.3) Asian: OR 2.1 (95% CI: 1.4-2.7) Black: OR 3.0 (95% CI: 1.0-5.0), ..... etc. I would like to adjust these ORs by age (as a categorical variable), but when I run the code, I cannot figure out how to get ORs for each of the covariates listed above and not just the overall aOR. What I want is an output that shows all the adjusted ORs when adjusting for age among all of the covariates (as shown above: sex, race, etc.). Below is the code I'm currently using. I assume I need to add in a BY or SLICE statement, but both of those strategies don't seem correct: proc logistic data=disease; class Clinic (Ref = 'A') Sex Race / param=glm; model Disease (event = "Yes") = Clinic; oddsratio Clinic; run; Thanks.
... View more