The last time I did this was about 15 years ago, and I did it 'by hand' in a DATA step. So I read through the data, and counted number of incidents, and added up the number of person-years, then at the end I did some math for the confidence interval. That's definitely an option. If your data are well structured, it's pretty straight forward.
I would start by heading to lexjansen.com and searching the papers indexed there for "incidence".
I found a few with macros, but also saw that SAS/STAT 12.1 introduced PROC STDRATE, which looks like it will do it:
https://lexjansen.com/nesug/nesug12/sa/sa19.pdf
https://support.sas.com/resources/papers/proceedings13/423-2013.pdf
... View more