zero inflated negative binomial for a rate outcome

[LucyB]

I have an outcome frequency over a variable number of years per patient. I have converted this to a rate per year and would like to investigate differences bet a certain category. The rates are overdispersed and zero inflated. How can I code a zero inflated negative binomial model in SAS for this rate outcome?

[Ksharp]

PROC GENMOD

[LucyB]

what about countreg?

e.g:

proc countreg data=data;
    model rate = category covariate / dist=zinb method=qn;
    zeromodel rate ~ category covariate;
run;

I was not sure how to model the zeromodel..

[Ksharp]

I don't know PROC COUNTREG is under SAS/ETS, it is for time series.