Induced Abortions in Matlab, Bangladesh: Trends and Determinants

Abstract

This study examined the reasons for induced abortion among women in Matlab Bangladesh and the multiple determinants of abortion. Data were obtained from the longitudinal surveillance system among about 75000 pregnancies during 1982-91 in Matlab and comparison areas. Explanatory variables were age parity pregnancy interval contraceptive use prior to pregnancy maternal education household space study area and calendar year of pregnancy termination. Of the 33473 pregnancies during 1982-91 in the treatment area 30110 were live births 1689 were miscarriages 1064 were stillbirths and 610 were induced abortions. The abortion ratio was 20/1000 live births in the treatment area and 33/1000 in the comparison area. Abortion ratios increased over time in both areas. The induced abortion ratio increased with maternal age and parity. The risk of induced abortion was much higher if the former pregnancy interval was under 12 months. Likelihood of induced abortion increased with maternal education and household space in both areas. The induced abortion ratio in the treatment area among contraceptive users in the former pregnancy were twice that of nonusers. Proportional hazards models indicate that abortion was more common among women with higher parity shorter pregnancy intervals more education and higher socioeconomic status. During 1992-95 induced abortion and miscarriage declined in the treatment area. Higher ratios among users of some contraceptive methods suggests a link with use-failure rates. The lower level of abortion in the treatment area suggests successful maternal-child health and family planning programs.