Can anyone help me with finding the CRLB for unkown p for a NB distribution?

I have a random sample of Xi's from NB(r,p) where p is my parameter of interest, r is unknown but has a value.

I know the CRLB formula to use, I'm just stuck on substituting in the Variance for NB. I know the variance is rq/p^2 for NB.

I've gotten to the point where

CRLB= 1/ nE({r-px}/p{1-p})^2

Please help.