topic incident rate while controlling for confounders in SAS Programming

incident rate while controlling for confounders

katemcguckin — Sat, 10 Aug 2019 11:45:28 GMT

Hi,

Say you have an incidence rate in PYs.

Suppose the variable is called MENNO = number of cases of meningitis

And you have another variable called PYTOTAL = total person years of study

1. How do you program, in SAS, the incidence rate in person-years? (IR in PYs)

2. How do you program a 95% CI of the IR in PYs?

3. How do you program the IR in PYs after confounding, say for age, sex, race?

I don't have any data yet. But I NEED to know the programming steps for these 3 questions and can't find them on the internet.

I suck at macros at this point, too.

Please help!

Kate

Re: incident rate while controlling for confounders

PaigeMiller — Sat, 10 Aug 2019 11:50:21 GMT

It's hard to give code without being able to reference an actual data set.

But, for example, take this part of your question

1. How do you program, in SAS, the incidence rate in person-years? (IR in PYs)

If you had to do this without a computer, in other words, do it in your head, or pencil and paper, what are the calculations?

Re: incident rate while controlling for confounders

Reeza — Sat, 10 Aug 2019 17:41:15 GMT

Isn't that just MEMNO/PYTOTAL?

If you don't have real data, make fake data and work with that. Working with imaginative constructs is much harder, especially trying to explain it to someone else.