Child loss and its impact on fertility.

This paper describes the impact of infant and child mortality on birth spacing and on ultimate parity in Egypt. The data come from the 1980 Egyptian Fertility Survey. An infant death was shown to be associated with a reduction of about 6 months roughly 15-20% in the interval to the next birth. Pronounced behavioral and biological mechanisms were shown to account for the reduction. There was a significantly lower likelihood of contraceptive use in the event of an infant death which persisted after taking parity and socioeconomic status into account. A significant reduction in interval length associated with an infant death was observed among users and non-users of contraception though it was more pronounced among the former. Most of the reduction in interval length among non-users was accounted for by the biological effect of the reduction in the anovulatory period. The persistence of a mortality effect among non-users as well as users after controlling for the duration of amenorrhea was indicative of some behavioral response also among the former. Cairo Alexandria and urban Lower Egypt featured proportionately the largest behavioral response while the largest biological effect was found in Upper Egypt. There was a pronounced attitudinal replacement response to the number of child deaths within each domain of residence. On the whole the response fell short of total replacement. At the individual level the death of a child denotes both a significantly greater likelihood of desiring another birth promoted largely by a replacement response and also a significant reduction in the interval to the next birth.