About half of maternal deaths in the world in 2005 occurred in Ethiopia, Nigeria, Democratic Republic of Congo, India and Afghanistan . In Ethiopia, the maternal mortality ratio (MMR) was 720 per 100,000 live births in 2005, and about 90 times higher than in resource-rich countries. Reducing the maternal and neonatal mortality are the targets of the Millennium Development Goals 4 and 5 .
As it is true for most of Africa, Ethiopia lacks information for setting priorities and monitoring health interventions. Unfortunately, it is difficult to get reliable measures to follow the progress on interventions to reduce maternal and neonatal deaths. Some of the tools used to estimate maternal and neonatal mortality rates provide temporary estimates, but are not so helpful in overseeing progress of interventions [3, 4]. There are few studies in Ethiopia that compares measures as the sisterhood method with population-based birth and death registration. In addition, our study represents an opportunity to link information between communities and health institutions.
This research is linked to an intervention programme (Reducing Maternal Mortality; see http://www.lindtjorn.no/page1/page11/page11.html) in south-west Ethiopia. Our aim is that this research will provide us with the tools to measure the impact of interventions.
We aim to set up a simple and sustainable birth registry to oversee maternal and newborn deaths in a remote part of Ethiopia.
The overall aim of the study is to develop tools to monitor maternal and neonatal deaths in communities and health institutions in south Ethiopia, and to estimate effective coverage of emergency obstetric care.
Through this research we will estimate of the size of the problem, identify causes of maternal and neonatal deaths, and enable regular monitoring of progress.
These research objectives are:
- To estimate maternal mortality ratio and identify risk factors for death
- To determine neonatal mortality rate and identify risk factors for death
- To estimate pregnancy-related deaths in the area using the direct sisterhood method
- To assess the availability and performances of emergency obstetric care at institutions by reviewing the birth registries at hospitals and health centres.
- To monitor trends in maternal and neonatal mortality rates by using community and institutional birth registries
- To compare the results of maternal and neonatal death estimates by using population based registries, institutional registries and direct demographic methods.
- To estimate effective coverage of emergency obstetric care
Expected outcomes of the research
Through this research we will get an estimate of the size of the problem, identify causes, and determinants, identify differences in levels within our catchment area and the research would enable regular monitoring of progress. It is a model for Ethiopia, and may also help other countries to set up similar registries.
We shall use and compare several designs to measure maternal and neonatal mortality:
- Population based registry: The Health Extension programme in focuses on preventive and promotive health care. Two women with one year training work as health extension workers in their kebeles (average population 5000 people). They receive a salary from the government, and work on disease prevention and health promotion conducted by house-to-house visits. As the HEWs regularly visit homes, they will learn about all births in their catchment area. We shall train HEWs to register births and deaths, and find out the possible causes of death using verbal autopsies. The population-based birth and death registry (se uploaded copy of registry form) is a prospective cohort to follow the outcomes of about 23,000 live births in four districts (Dirashe, Bonke, Arba Minch Zuria and Demba Gofa with a population of 592.000 people). The outcome measures are maternal deaths within 42 days, and death of newborn babies within 28 days after delivery. Here you can download The RMM-Birth-registry-form and the its English translation
- Using the sisterhood and household death survey method  we shall collect data from Bonke district (population 173,000) to estimate maternal deaths in the last 5 years.
- Institutional based birth registry: In this retrospective review of 68 health institutions in Gamo Gofa Zone (population 1,6 million people), we assess the standard and quality of obstetric care [5, 6], and record outcomes such as maternal and neonatal deaths, and birth complications.
- The effective coverage framework points out what fraction of the potential health gain achievable through the interventions is being delivered. To estimate the effective coverage of emergency obstetric care, we shall use the method proposed by Shengelia and others 
This research does not contain any new intervention or drugs. Women in need of health care will be referred to the nearest institution. The study will be conducted according to the Declaration of Helsinki, and we shall get ethical clearance from the Ethical Review Board of the Southern Regional Health Bureau in Ethiopia. All study records will be maintained in a secured location.
We shall engage key stakeholders through meetings and workshops to ensure the research is relevant to the context. Communications approaches include: community groups; workshops; and meetings with key individuals. We shall tailor all communication to meet the needs of the groups and that messages are clearly communicated.
For each sub-objective, we shall publish papers in international peer-reviewed and open – access journals, and we shall present reports to central and regional health authorities in Ethiopia. We shall use the results from the research in teaching and training activities at UIB and Ethiopian universities.
The research is collaboration between Ministry of Health in Ethiopia, Arba Minch University, Arba Minch Hospital, Gidole Hospital and University of Bergen.
Yaliso Yaya (BSc, MPhil), Arba Minch University and Arba Minch College of Health Sciences, PhD student (Quota scholarship).
Tadesse Data (BSc), Gamo Gofa Zone Health Department, MPH student at Gondar University, works on the birth and death registries.
Meseret Girma (BSc), Arba Minch University, MPH student at Gondar University, works on the health institution.
Professor Bernt Lindtjørn, UoB: Coordinator and main supervisor
Professor Ole Frithjof Norheim, UoB, co-supervisor