Factors associated with adolescent abortion in a rural area of Bangladesh.

This study examines levels and trends in abortion among adolescent girls in 1982-98 and identifies groups of adolescents who are at high risk for having an abortion. The study used data sets collected in Matlab in Bangladesh where the Center for Health and Population Research at ICDDR,B, has maintained a demographic surveillance system, since 1966. Both bivariate and multivariate techniques of analysis were employed. We also used qualitative information derived from in-depth interviews with adolescents. The incidence of abortion was 35 times higher for unmarried than for married adolescents. Abortion ratios were also higher for adolescents who were <18 years old and for those with more than primary education. A little less than half of the abortions were induced by biomedical health workers (by means of menstrual regulation) which means that traditional providers fulfilled an important function in this rural area. The findings suggest that high-quality abortion services provided to adolescents in the framework of comprehensive reproductive health services will help to lower abortions in general and of unsafe abortions provided by traditional providers in particular.