A Model for Studying Birth Rates Given Time Dependent Changes in Reproductive Parameters

Abstract

If the reproductive behavior of a population changes radically, this change will be reflected in birth rates only after a time lag. In addition, however, an observed initial response to change (after a lag) is often followed by a partial regression toward the previous level. Is such a regression due to another change in the patterns of reproductive behavior, or may it be the built-in consequence of the original change in behavior, as the rates approach a new equilibrium? A time-dependent reproductive model, based on biological considerations, permits investigation of fluctuations in birth rates as a consequence of fluctuations in the probability of being able to conceive at any given point in time, i.e. of being susceptible to conception. Early effects of a change in behavior are likely to assume an extreme value that will not be seen again barring another change in behavior. Accordingly, differentials and trends in birth rates do not necessarily indicate corresponding relationships in the probability of conceiving or in abortion rates. Inferences from levels of birth rates about levels of these underlying biological factors should be made with caution. In particular, the built-in oscillations in the probability of being susceptible to conception, and the resulting oscillations in birth rates must be considered.