A new study published this week in the medical and public health journal The Lancet indicates a drop in maternal deaths worldwide for the first time in 30 years, from roughly 525,000 in 1980 to about 343,000 in 2008.  Overall, the study suggests an estimated yearly rate of decline in maternal deaths of 1.5 percent.

Maternal mortality and morbidity–from causes related to pregnancy, delivery, usafe abortion or childbirth up to 42 days after delivery–have been the leading causes of death and illness among women ages 15 to 49 in many countries throughout the world. Efforts to reduce maternal deaths and illnesses have been a focal point of international agreements for decades ranging from Alma Ata Declaration to the International Conference on Population and Development to the Millenium Development Goals, and a priority for women’s rights and health groups throughout the world.

As the study notes, estimated reductions in maternal mortality have not occurred consistently across all nations and in fact have risen in some countries. Despite the encouraging findings, for example, the study notes “a surprising result [in] the apparent rise in the MMR [maternal mortality rate] in the USA, Canada, and Norway.”

Complications of unsafe abortion are a major contributing factor to maternal death and illness worldwide, responsible for roughly one-fourth of all maternal deaths. Curiously, the new study never once mentions the word “abortion.”

A Rapidly Changing Picture?

Maternal mortality and morbidity (death and illness) are notoriously difficult to track for many reasons, including the lack of accurate records, misidentification of cause of death, stigma and problems related to tracking illness and death among widely dispersed populations not well-served by accessible health services.  And, for a long time, maternal death was considered, socially, culturally, politically and economically as being “not very important,” or “a given,” attitudes that women’s groups have worked for decades to change.

Other recent estimates of deaths from pregnancy and childbirth, including another analysis published in The Lancet in 2007 had indicated little progress in reducing maternal deaths during the past couple of decades.  The 2007 analysis, for example, conducted by researchers from the United Nations Children’s Fund (Unicef), Harvard, the World Bank, the World Health Organization (WHO) and the Johns Hopkins School of Public Health, estimated roughly 535,900 maternal deaths in 2005, the figure still in use  by WHO.

The new analysis was conducted by Dr. Christopher Murray, the director of the Institute for Health Metrics and Evaluation at the University of Washington, in Seattle in collaboration with researchers at the School of Population Health, University of Queensland, Australia. The Institute for Health Metrics and Evaluation was established with a $105 million gift from the Bill & Melinda Gates Foundation and $20 million from the state of Washington. The Gates Foundation supported the research for the maternal mortality study.

That the new study will draw scrutiny is assumed, especially in light of both the challenges of measuring maternal mortality, the uncertainties of available data, and the dramatic changes implied by the analysis.  As noted by Richard Horton, Editor of The Lancet, who calls for the study to galvanize further research and analysis:

[T]hese results will provoke intense debate among the global health measurement community. This much was clear during the peer review process. For example, although reviewers concluded that this study was “well designed…[and] well explained”, “a very important contribution to our understanding of the epidemiology of maternal mortality”, and “seeks to make a crucially important contribution to the global monitoring of maternal mortality”, concerns about uncertainty estimates, in particular, were common. Understanding the varying approaches to measuring maternal mortality—their strengths, weaknesses, advantages, and disadvantages—must now be a priority for all those concerned with translating global health numbers and country estimates into policy.

I note again with curiosity that Horton’s editorial never once mentions abortion.

Driving the Trends

The study points to “[f]our powerful drivers of maternal mortality [that] are improving in most countries.”

One is a drop in total fertility rates (TFR), as desired family size falls in many countries. The global TFR, or average number of children born per woman of childbearing age, has dropped from 3.7 children per family in 1980 to 2.56 in 2008, according to the authors, so “[d]espite rising numbers of women of reproductive age, the decrease in TFR has kept the size of the global birth cohort stable.”

Smaller desired family size along with later age-at-marriage, increased to access to contraception and safe abortion services, or reliance on unsafe abortion, have led to declining birth rates in countries throughout Asia, Central Europe and Latin America, and have been important contributors to reductions in maternal death rates in those regions.  A major factor in these declines are the reduced total fertility rates in such highly populous countries as India and China, and in other countries of Asia. 

Increased per capita income is another factor.  Changes in income, note the authors, “can affect maternal mortality through several channels from nutritional status of mothers to physical and financial access to health care.”

Third is education. As educational rates among women rise, they are more likely to want smaller families and know more about contraception and the means to prevent unintended pregnancy, have better incomes, and have better access to both preventive and curative health care.  And fourth, the authors speculate, is increased access to skilled birth attendants. “[A]lthough [our model] did not include the proportion of women giving birth with a skilled attendant, the steady, albeit slow, rise in coverage of skilled birth attendance could have contributed to maternal mortality declines.”  They cite the fact that some large countries such as India have witnessed quite rapid increases in skilled birth attendance in recent years as a factor.

Increased access to safe abortion services?  No mention.  In fact, there is no assessment here either of expansion of access to emergency obstetric care, which might also be playing a role in some countries where such care has long been inadequate to say the least.

Scattered Progress

According to this study, the “burden” of maternal mortality has shifted. For example, as the share of maternal deaths worldwide contributed by Asia has fallen, the share occurring in sub-Saharan Africa has increased. The authors write:

The number of births globally varied between 124 million in 1980 and 136 million in 2008. The regional composition shifted slowly towards sub-Saharan Africa, with the most noticeable change being the reduction in the number of births in east Asia. The proportion of global maternal deaths in sub-Saharan Africa increased from 23% (18—27) in 1980 to 52% (45—59) in 2008, resulting from both the accelerated increase in the number of maternal deaths in the early 1990s and declines in Asia.

In addition, the authors note, a small number of countries still account for a large share of total global maternal deaths, either because the rates of maternal death have not declined as fast as in other countries, or because large populations result in higher overall numbers of maternal deaths despite declining rates.  Providing a number of tables and charts to underscore their findings, the authors note:

The webappendix p 9 shows the 21 countries with the highest numbers of maternal deaths in 2008. Together, these countries represent 79·4% of total global maternal deaths and 60·6% of global livebirths. Although table 3 shows that MMRs were substantially higher in sub-Saharan Africa than in other regions, south Asia was a major contributor in terms of total numbers of maternal deaths.

Global maternal mortality rates might have fallen faster, according to the authors, but for one factor: the global AIDS epidemic. High rates of HIV infection and AIDS-related illness among pregnant women have contributed to higher rates of maternal mortality, especially in parts of sub-Saharan Africa. The researchers estimated that absent HIV, the annual rate of decline in maternal deaths according to their data would have been closer to 2.2 percent.

Declining rates of death and illness in some countries also come as the result of investments in basic health care.  A study by the Partnership for Maternal, Newborn and Child Health notes that the government of India has in place a plan that pays women to get prenatal care and skilled care for delivery. Nepal provides home visits for family planning. And Malawi is training nonphysicians to perform emergency Cesarean sections, a critical intervention in places where well-equipped clinics are few and far between and where obstructed labor and other related complications can spell death.

Moreover, some countries such as India and Nepal have made a commitment, at least on paper, to reduce maternal death and illness related to complications of unsafe abortion. Abortion is legal in India, and Nepal legalized abortion in 2002, and increased access to safe services is unquestionably a factor in reduced rates of maternal mortality in these countries. Still, a large share of women in both countries face obstacles to obtaining safe abortion services due to inconsistent investment in these and other basic health care services. 

Meanwhile and surprisingly, or not if you know more about birthing in the United States, the U.S. joins Canada, Norway, and Afghanistan as countries in which maternal mortality rates actually rose, although the authors cite delays in reporting of data as one possible reason for the anomalies in the U.S., Canada, and Norway.

Moving Forward

In his editorial, Horton recounts pressure he received from advocacy groups requesting delay of the release of this study, for fear it would undercut efforts to prioritize women’s health at upcoming meetings, such as the G8 Summit this June in Canada, the MDG Summit in September, and the Women Deliver Conference also this June in Washington, DC, or that the findings would precipitate reduced urgency of funding for addressing maternal mortality by the United States and other major donors.

Even before the paper by Hogan et al was submitted to us, we were invited to “delay” or “hold” publication. The justification for this concern was several fold: potential political damage to maternal advocacy campaigns; confusion among countries, policymakers, and the media, given the difference between this maternal mortality estimate and the previous UN number; undermining progress on global commitments to maternal health; and the risk of an unproductive academic debate while women continued to die. Although well-intentioned, these requests to slow the pace of scientific discussion for political considerations are likely to be far more damaging than fostering a serious debate about progress in reducing maternal mortality as and when new data appear. Is the global health community unable to accommodate diverse voices and sources of evidence? Is it unable to create constructive ways to bring scientists and policymakers together to reach agreement about the meaning of new research findings?

[A] process needs to be put in place urgently to discuss these figures, their implications, and the actions, global and in country, that should follow. Ban Ki-moon is currently leading a Joint Effort on Women’s and Children’s Health. The purpose is to plan measures for the MDG Summit in September. One outcome of the UN Secretary-General’s important initiative might be to convene a high-level, intergovernmental MDG preparatory meeting as a satellite event at Women Deliver in June. The goal of such a preparatory meeting would be to bring the best available data to bear on formulating policies to launch in September.

In fact, all of this is true.  We need the best data available to best assess needs and allocate resources.  We need to embrace success and fight for our priorities.  We need to work to address maternal illness and death but place this in the context–unrelentingly–of the broader reproductive and sexual health needs of women whether or not they are pregnant.  And we need to secure access to safe abortion services for all women, everywhere, so that every pregnancy is a wanted pregnancy and every child a wanted child.

In short, we need to ensure that all of the indivisible issues of women’s rights and agency around whether, with whom and when to bear children, and surrounding their rights to access to all relevant sexual and reproductive health services make up the broader vision in which we focus our specific efforts, never allowing ourselves, no matter who we are or where we sit, to be silent on “difficult” issues, nor suppress our own voices under political or other pressures.