Journal Article

Reviewing maternal mortality in rural Ethiopia: Using the verbal autopsy approach

Background
Maternal mortality and morbidity in Ethiopia are among the highest in the world and stem from a range of socio-economic, political and demographic factors. Verbal autopsy is one approach used to review maternal deaths in settings where hospital-based audits and confidential enquires are not possible.

Objectives
This study aimed at analyzing the cultural, social, economic, behavioral and biological factors that influence maternal mortality in Ethiopia.

Methods
In this study, verbal autopsies were employed as part of a larger community-based safe motherhood study which gathered qualitative data from communities distributed across Ethiopia’s eleven regions in 2005. Following discussions with community members about recent maternal deaths, researchers contacted the families and requested an interview with them. In total seventeen verbal autopsies on maternal deaths were carried out. Verbal autopsy is a method used to record events leading up to the cause of death. It is widely used in countries where vital registration and death certification systems are weak and most women die at home without medical certification of the cause fo death.

Results
Of the seventeen verbal autopsies, seven were recorded from Southern Nationalities Nations and Peoples Region (SNNPR); four were recorded from Tigray and three from both Oromiya and Amhara Regions. Ages of women who died ranged from 20 to 49. All women except one were married. Just over half of the women (nine) attended antenatal care in a health facility and perceived that because there were no problems during pregnancy, they could deliver at home. The remaining seven did not attend any health care during pregnancy. Most women (12) gave birth at home; three on the way to the facility, two in the facility (although one failed to deliver before she died). Sixteen out of the 17 women who died started labour at home, and the delivery was initially attended by female relatives or TBAs. Half of the women (9) were delivered by a relative, six by a TBA and two by a nurse midwife. Eight of the women who died were said to have been in poor health prior to delivery. Majority of women were multiparious: six women had six pregnancies, and four women had 3-5 pregnancies. Two were giving birth to their second child. Only two young women died as a result of their first pregnancy. Five of the women died almost immediately after giving birth.

Conclusion
Although these verbal autopsies failed to identify the specific medical causes of maternal deaths, they clearly show that women continue to have limited access to emergency obstetric care. Single solutions such as the promotion of ANC and use of TBAs have not resulted in reductions in maternal mortality. Although the Health Extension Program is underway more is required to save mother’s lives such as ensuring skilled birth attendance and swift referral for EmOC services to achieve better results for mothers and newborns in the country.