Safe Motherhood
Maternal Mortality Measurement


A brief history of safe motherhood

In 1985 Rosenfield and Maine wrote a highly influential paper that galvanised interest and put the issue of maternal mortality on the international health policy agenda:

The Lancet

They argued that maternal and child health (MCH) programmes focused almost exclusively on child health, assuming that “whatever is good for the child is good for the mother”.

The first international conference devoted to maternal mortality (Safe Motherhood Conference, Nairobi, Kenya, 10-13 February 1987) was sponsored by the World Bank, WHO, and UNFPA and led to the launch of the Safe Motherhood Initiative (SMI).

International agencies involved in the SMI coalition included five UN agencies (WHO, UNDP, World Bank, UNFPA, and UNICEF) and two NGOs (the Population Council and IPPF). Family Care International, another NGO, also came to be involved in organising the first national conferences on safe motherhood and was the secretariat for the programme.

The SMI then broadened to include other groups. By 2005 it was subsumed as part of the Partnership for Maternal, Newborn and Child Health (PMNCH) with a much expanded scope centred on alliance between the maternal, newborn and child health communities.

Back to Top

The history of maternal mortality measurement

The history of measuring maternal mortality - “deaths in childbed” - goes back over two centuries in some countries:

Oxford University Press

In most cases, the measurement evolved with the overall civil registration system, but in some settings, maternal deaths were initially recorded as part of the maternity service, with health care providers primarily responsible for reporting cases:

Cambridge Journals Van Poppel F, Van Dijk JP (1997) The development of cause-of-death registration in the Netherlands, 1865±1955.Continuity and Change;12: 265-87

As experience of measuring maternal mortality increased, so did reliance on multiple data sources and capture mechanisms. This evolution was partly as a consequence of the increasing rarity of maternal deaths and the need for intensified surveillance so as not to miss cases. It also reflected a realization that all measurement approaches have advantages and disadvantages, both in terms of complete ascertainment of maternal deaths and provision of additional information for programmatic purposes.

Today in many countries with advanced statistical systems, maternal deaths are still identified from multiple sources, as for example in the UK Confidential Enquiries into Maternal Deaths (CEMD):

Royal College of Obstetricians and Gynaecologists

and with additional special efforts, such as inclusion of a pregnancy-status check box on death certificates as well as active surveillance in the United States:

Elsevier - American Journal of Preventive Medicine MacKay AP, Rochat R, Smith SC, Berg CJ. The check box. Determining pregnancy status to improve maternal mortality surveillance. Am J Prev Med 2000; 19: 35-39.

In developing countries, heightened interest in maternal mortality coincided with a series of sub-national special studies in the 1980s that revealed a higher than expected frequency of maternal deaths:

The World Health Organisation WHO. Maternal Mortality Ratios and Rates: A Tabulation of Available Information. Geneva: World Health Organization. 1991

These dedicated studies also highlighted the serious underreporting in routine statistics and gave early insights into the challenges of capturing maternal deaths, particularly where the vast majority occur at home without contacting the health system.

At this time, the options for measuring maternal mortality at a population level in developing countries were essentially limited to those used for all adult mortality - via incomplete vital registration or large-scale population surveys.

Alternatively, deaths to women of reproductive age were adjusted by the proportion estimated to be due to maternal causes - often assumed to be 25-33% - to arrive at estimates of maternal mortality.

World Health Organisation AbouZahr C. Maternal mortality overview. In: Health Dimensions of Sex and Reproduction. Global Burden of Disease and Injury Series. Murray CJ, Lopez AD (eds).Geneva: World Health Organization, Harvard School of Public Health, World Bank
Harvard School of Public Health
World Bank

This type of approximation process continues to be used today, as seen for example in the UN model estimates

The Lancet Hill, K et al . Estimates of maternal mortality for 2005. Lancet 2007 (forthcoming)

Since then the measurement opportunities for maternal mortality have increased markedly. Some reflect wider developments in information gathering in developing countries, such as the Demographic and Health Surveys or sample vital registration, or capacity-strengthening initiatives such as the Health Metrics Network. Other opportunities reflect measurement advances, for instance:

Studies in Family Planning Graham WJ, Brass W, Snow RW. Estimating maternal mortality: the sisterhood method. Studies in Family Planning 1989; 20: 125-135
Immpact Immpact Toolkit: a guide & tools for maternal mortality programme assessment: Module 4: Sampling at Service Sites (SSS)
Bulletin of the World Health Organisation Stanton C, Hobcraft J, Hill K, et al. Every death counts: measurement of maternal mortality via a census. Bulletin of the World Health Organization 2001; 79: 657-664
Amstrup SC, McDonald TL, Manly BFJ (2005) Handbook of Capture-Recapture Analysis: Princeton University Press

The crucial point is that for most situations and purposes, there is now an opportunity and a method suitable, if adequate resources are provided.

The diversity of options is partly a consequence of the diversity of measurement contexts in developing countries, and partly a response to the challenges of capturing maternal deaths. Explicit recognition of the problems of measurement is necessary both to stimulate and inform the search for solutions.

The decade up to 2015, the reporting year for the Millennium Development Goals, could witness great progress in measurement if more opportunities are realized.

Back to Top