The descriptive epidemiologic measures are incidence rate, cumulative incidence and prevalence. This chapter explains how the precision of an observed value, can be determined. Analyses of incidence rate are generally based on the assumption that the number of cases has a Poisson distribution. A precondition for using this model is that the number of cases is seen as an observation of a stochastic variable while the time at risk is understood as a constant. It is possible to establish probability models where the time at risk is also seen as a stochastic variable. The total time at risk, summed for all individuals, is understood as an interval of time during which a case of illness could occur at any point in time. The binomial distribution is generally the starting point for analyzing the cumulative incidence. An alternative approximation process can often be used in epidemiology because of the low frequency of many diseases.