Pregnancy termination in Matlab, Bangladesh: maternal mortality risks associated with menstrual regulation and abortion

Abstract

Context: In Bangladesh, both menstrual regulation (MR), which is thought to be a relatively safe method, and abortion, which in this setting is often performed using unsafe methods, are used to terminate pregnancies (known or suspected). However, little is known about changes over time in the use of these methods or their relative mortality risks.

Methods: Data from the Demographic Surveillance System in Matlab, Bangladesh, on 110,152 pregnancy outcomes between 1989 and 2008 were used to assess changes in mortality risks associated with MR (and a small number of dilation and curettage procedures), abortion and live birth. Tabulation and logistic regression analyses were used to compare outcomes in two areas of Matlab--the comparison area, which receives standard government health and family planning services, and the Maternal and Child Health-Family Planning (MCH-FP) area, which receives enhanced health and family planning services.

Results: In Matlab as a whole, the proportion of pregnancies ending in MR increased from 1.9% in 1989-1999 to 4.2% in 2000-2008, while the proportion ending in abortion decreased from 1.6% to 1.1%. The odds of mortality from MR were 4.1 times those from live birth in 1989-1999, but were no longer elevated in 2000-2008. The odds of mortality from abortion were 12.0 and 4.9 times those of live birth in 1989-1999 and 2000-2008, respectively. Reduction in mortality risk was greater in the MCH-FP area than the comparison area (90% vs. 75%).

Conclusion: MR is no longer associated with higher mortality risk than live birth in Bangladesh, but abortion is.