Relationship among Contraception, Abortion, and Fertility: Lessons Learned from Matlab, Bangladesh

Abortion was found to increase with contraception in some populations and decrease in other populations. The study investigated the reasons for this contradiction using selective measures limited by paucity of detailed data on abortion from the Matlab Demographic Surveillance System (DSS), Bangladesh, for the 1978-1998 period. A comparison of abortion, contraceptive prevalence rate (CPR), desired fertility, and total fertility rate (TFR) over time formed the basis of the study. The desired fertility declined over time. Fertility was found to be converging to the desired fertility, and the process of convergence was faster in the MCH-FP area. The relationship between the CPR and the total abortion rate (TAR) was positive, negative, and zero during the convergence. Magnitude of abortion depends on the quality of reproductive health services; and a comprehensive MCH-FP program is expected to bring down both fertility and abortions substantially by increasing contraception. Both fertility and abortion are problems in Bangladesh and many other developing countries. About 50 million abortions occur each year in the world, of which about 25 million are illegal (Henshaw, Singh and Has, 1999). In Bangladesh, maternal death (about 4 per 1,000 live births) is one of the highest in the world; about 25 percent of these deaths occur due to abortion (Mohammad, 1986; Fauveau and Blanchet, 1989). Abortion is also a leading cause of maternal morbidity in Bangladesh (Khan et al., 1986). It is, therefore, important for Bangladesh and many other developing countries to know how both fertility and abortion can be reduced. Abortion was found to increase with contraception in some populations, but decrease with contraception in some other populations. It was also found to increase in a population for a period, but decrease in the same population for a different period (Singh and Sedgh, 1997; Alan Guttmacher Institute, 1999; Senlet, Curtis, Mathis, and Raggers, 2001). This