Measuring maternal mortality in southern countries: an example from Cameroon

Maternal mortality is defined as death during pregnancy (at any stage of pregnancy) or within 42 days of delivery from any cause related to pregnancy or its care. The mortality ratio is measured by comparing the number of maternal deaths to 100,000 live births in a given year.

[Un article issu de The Conversation écrit par Marion Ravit, Docteure en Santé Publique, Chargée de Recherche, Institut de recherche pour le développement (IRD), Benjamin Fomba Kamga, Enseignant-chercheur, Université de Yaoundé II et Martine Audibert, Directrice de recherche émérite, Centre d’Etudes et de Recherches sur le Développement International, CNRS, UCA]

Map of maternal mortality ratio in 2020

(maternal mortality per 100,000 live births)

Maternal mortality difficult to measure in the South

While maternal mortality is low and very well estimated in Europe (less than 10 per 100,000), the incompleteness of civil status registers and limited access of pregnant women to health establishments pose problems for its measurement in the South.

If we take the example of Cameroon, it is estimated that nearly two thirds of children are not registered at birth, which also has consequences on the estimation of neonatal mortality.

In addition, a third of Cameroonian women give birth at home (according to the latest available figures dating back to 2018), which makes it impossible to know how many of them died as a result of their pregnancy or childbirth.

Sisters' Method and Other Alternative Measuring Tools

There are alternative methods for measuring maternal mortality, including:

• the sisters method: this method is the most widely used to date in low- and middle-income countries by international organizations and researchers to estimate maternal mortality; we detail it further below.
• RAMOS Surveys (Surveys on Reproductive Age and Mortality): they aim to identify maternal deaths that occur within health institutions and in the population, by combining different existing data sources. One of the limitations lies in the quality of the data sources;
• demographic surveillance studies: they are expensive, which limits their national exhaustiveness;
• studies from medical registers: they only record deaths at the institutional level while the rates of births outside the health system remain high, as mentioned above.

Let's go back to the sisters' method. It appeals to the memory of the women surveyed, while trying to minimize the risks of forgetting and errors in dating events. It consists of asking each woman over the age of 15 interviewed for the list and birth order of all the girls born to her biological mother, then asking if each of the sisters is still alive at the time of the survey.

For deceased sisters, age at death and number of years since death are asked. For those who died at age 12 or older, additional questions are asked to determine whether the death was potentially related to pregnancy, childbirth, or postpartum.

In Cameroon, assessing the effectiveness of the health check

In Cameroon, maternal mortality remains a major public health problem. It stood at 438 per 100,000 live births in 2020. It is crucial to put in place programs to reduce this maternal mortality and to have tools to measure their effectiveness.

Thus, in 2015, the Cameroonian government, with the support of the French Development Agency (AFD), implemented a system called “health check” in the three regions of the Far North of the country which are most affected by maternal mortality.

The health check is a prepayment mechanism for care for pregnant women to promote their access to this care. It aims to reduce maternal and neonatal mortality by improving the care of pregnant women.

The pregnant woman buys the health check for 6,000 CFA francs (around €10) and can then benefit from a package of care in accredited health facilities.

The health check includes: access to at least four prenatal consultations during pregnancy, coverage of pregnancy-related care, childbirth in an accredited health facility or transfer of the pregnant woman, in the event of complications, to the reference establishment and postnatal monitoring up to 42 days after childbirth.

It is essential to be able to estimate maternal mortality to know whether the health voucher has actually had an impact on it.

Improving estimates by crossing approaches

Maternal mortality, although still too high, remains a rare event that requires studying a large sample of women to detect its effects.

Our team therefore calculated that 45,000 women would have had to be surveyed. For budgetary and time reasons, such a large survey was not feasible. We therefore chose to turn to existing data from population-based household surveys: the Demographic and Health Surveys (DHS).

The latest DHS survey conducted in Cameroon in 2018 includes questions relating to the course of pregnancy and childbirth that are asked of women who had a live-born child in the five years preceding the survey. Since deceased women could not be interviewed, the sisterhood method was used to estimate maternal mortality up to seven years before the survey.

Although this method increases the potential number of maternal deaths observed, it remains insufficient to correctly estimate maternal mortality. Indeed, 14,000 women were interviewed and only 110 maternal deaths were recorded between 2012 and 2018 (between 1 and 24 deaths depending on the region).

Furthermore, one may wonder about the quality of the data collected. It is very likely that the women interviewed, who report the fate of their sisters following a pregnancy up to seven years ago, are mistaken about the year and cause of death. It is also possible that the death of a woman is counted twice if several sisters are interviewed.

Meeting the challenge of digitalizing civil status services

Currently, data from demographic and health surveys do not provide a reliable estimate of maternal mortality in countries where they are the most comprehensive and widely used. As long as indirect methods are used, the estimate of maternal mortality will remain approximate.

In the digital age, it is time to develop civil registration statistics systems to comprehensively record births and deaths in low-income countries. This is the intention of the National Bureau of Civil Status (BUNEC) of Cameroon, which is experimenting with the digitalization of civil registration services in some municipalities in the far north of the country.

The aim of this process is to facilitate the registration of births, making it possible to obtain a birth certificate in less than 24 hours and to declare the birth directly in a health facility.

However, for this system to be able to be extended throughout the country, many challenges still need to be resolved, such as power cuts and the lack of qualified personnel.