The reporting of maternal deaths and their causes varies widely and is associated with a country’s level of statistical development, although all countries tend to follow some version of the International Classification of Diseases (WHO, 1992). In countries with well-developed health and statistical reporting systems, the source of this information is the vital registration system. Separate cause-specific rates can be calculated for each of the major indirect causes of maternal death. The number of maternal deaths in a given facility or aggregate of facilities may be too small (e.g., fewer than 20) to calculate a stable rate for each complication. Therefore, in most facilities, only an aggregate indirect obstetric case fatality rate should be calculated. The data can be disaggregated by the type of types of facilities (e.g., basic versus comprehensive EmOC and/or public, private, non-governmental, community based) and by other relevant factors, such as district and urban/rural location.

Data Source(s):

National health information system; death records/certificates in the vital registration system; health facility records.

Purpose:

A substantial proportion of maternal deaths in most countries are due to indirect causes, particularly where HIV and other endemic infections, such as malaria and hepatitis are prevalent. Where infectious and communicable disease rates are high, often the number of maternal deaths due to direct causes is also high. With the rising prevalence of overweight among girls and women of reproductive age in many regions of the world, related chronic conditions, such as cardiovascular disease and diabetes, are increasingly contributing to indirect maternal deaths. This indicator does not lend itself easily to a recommended maximum level. Rather it highlights the larger social and medical context of a country or region and has implications for intervention strategies beyond EmOC where indirect causes kill many women of reproductive age. Research is needed in the area of indirect maternal deaths, including on the mechanisms by which indirect conditions cause maternal death and what programs could reduce them. This relatively new indicator of maternal mortality can be a proxy for the coverage and quality of programs for preventing or treating conditions associated with indirect maternal deaths (WHO et al., 2010) and relates to achieving Millennium Development Goals #5. improve maternal health #4. reduce child mortality.

Issue(s):

The indirect maternal fatality rates do not take into account deaths outside the EmOC health system and are not generalizable to the wider population. In settings where many women give birth at home or outside facilities, this indicator may be subject to bias because a disproportionate number of maternal deaths in a facility result from women who come for treatment of complications. In addition, the numbers of deaths and indirect obstetric case fatality rates may increase when efforts are made to improve hospital services and more women come for treatment. On the other hand, the absence of maternal deaths might indicate that women with serious complications are not brought to facilities or are referred on. The absence of reported deaths can also suggest that deaths are not being reported. The causes of maternal deaths are often misclassified, for example, the death of an HIV-positive woman might be classified as due to AIDS even if it was due to a direct cause such as hemorrhage or sepsis. On the other hand, due to lack of testing and reporting, as well as the associated stigma, HIV infection might be an underreported cause of maternal death. Misclassification of cause of death can lead to serious under-recording and problems of attribution of cause (WHO et al., 2010). Death certificates may never be filled out, may fail to indicate whether pregnancy was a recent occurrence, or may list multiple causes of death but an underlying cause is not registered. Similar to the recording of obstetric complications, training staff to comply with national standards of death certificate completion can result in more accurate and complete recording. Conducting reviews of all deaths of women of reproductive age in facilities, especially those who do not die on the maternity ward, could lead to more complete recording. Data collection for this new indicator may be difficult, however, the WHO technical consultation considered that it would be useful for governments and international agencies. In a few years, the indicator will be reviewed to see if it is useful and whether it should be modified (WHO et al., 2010).

References:

WHO, UNFPA, UNICEF, AMDD, 2010, Monitoring Emergency Obstetric Care: A handbook, Geneva: WHO. http://whqlibdoc.who.int/publications/2009/9789241547734_eng.pdf