Computer modeling of human fertility: the impact of reproductive heterogeneity on measures of fertility.

As a result of the paucity of biological markers for both reproductive events and exposure to reproductive toxicants, it is likely that individual members of populations will be misclassified with respect to reproductive performance and xenobiotic exposure. A four-parameter computerized model of fertility (frequency of intercourse, male fecundity, female fecundity, and spontaneous abortion) was developed to explore the effect of misclassification of populations on several measures of fertility. The measures of fertility explored include cumulative percent pregnant, time to pregnancy, cycle specific fertility, and fecundability ratio. The cumulative percent pregnant and time to pregnancy appeared to be insensitive to small changes in reproductive competence. The fecundability ratio decreased in proportion to the size of population exposed to a reproductive toxicant. The three reproductive parameters, time to pregnancy, cumulative percent pregnant, and fecundability, do not appear useful as measures of reproductive heterogeneity in populations. The reproductive heterogeneity of a population was best defined by the change in cycle-specific fertility rate over the period of observation. These simulations suggest that the change in the cycle-specific fertility rate should be evaluated and comparable in all population groups over the period of observation to assure reproductive homogeneity.