In brief, the uncertainty in the age-specific death rate in each county depends on the number of deaths (numerator) (n) and population (denominator) (N), and can be characterized with a binomial parameter with an expected value of p = n/N and a variance of p × (1 − p)/N.