Wendy Graham, Immpact Principal Investigator, and Professor of Obstetric Epidemiology, University of Aberdeen
Dr. Julia Hussein, Ipact Scientific Director, and Immpact Senior Research Fellow
Cynthia Stanton, Immpact Researcher, and Assistant Professor, Johns Hopkins Bloomberg School of Public Health
Sophie Witter, Immpact/Ipact Health Economist
Each year, more than 500,000 women in the developing world die during pregnancy or childbirth. One in 16 women in Africa and 1 in 43 women in Asia will die of maternal causes this year, compared to 1 in 2,500 in the United States. In the next several months, the Initiative for Maternal Mortality Programme Assessment (Immpact) will complete a six-year program of global research that identifies and measures the effectiveness of strategies to reduce maternal mortality. This extensive program—funded by the Bill and Melinda Gates Foundation, the UK Department for International Development (DFID), the European Commission, and the U.S. Agency for International Development (USAID)—has developed new tools to robustly assess maternal health programs in a number of developing countries. On March 12, 2008, four members of the Immpact research team, along with a DFID representative, shared some of their key results and selected strategies for the first time with their North American colleagues.
This event follows two related events held at the Wilson Center. In January 2008, the Center hosted the U.S. launch of a new series in the British medical journal, The Lancet, that described the relationship between maternal and child undernutrition and outlined their short- and long-term consequences. The authors offered evidence-based interventions to significantly reduce the high morbidity and mortality rates caused by this global health burden. The Lancet series aimed to highlight the crucial role early nutrition plays in public health and economic development in order to stimulate improvements in funding, policy, and interventions. In October 2006, the Center hosted a launch of the Maternal Survival Series in The Lancet. This special issue called attention to the progress in—and challenges of—reducing maternal mortality in both the developing and the developed world, and outlined what the authors believed to be the "best bet" strategy for preventing these deaths.
A Slow Start
"One of the great tragedies of international development [is] the failure to make any impact on maternal health," said Stewart Tyson, head of DFID's Health Advisory Group. "The level of deaths are the same as they were twenty years ago…Despite the progress in many areas—AIDS, TB, malaria, child health—we've not made a dent in this." In spite of high visibility and considerable funding, efforts to reduce maternal mortality are off to a slow start. Millennium Development Goal (MDG) 5, to reduce by three-quarters the percentage of women dying in childbirth by 2015, is in danger of not being achieved. MDG 4, to reduce by two-thirds the mortality rate among children under five, is also showing slow progress.
Wendy Graham, a professor of obstetric epidemiology at the University of Aberdeen and the principal investigator for Immpact, introduced the strategy and methodology of this six-year project. Recognizing the complexity and global reach of the problem, Immpact brought together seven research partners from both the Global North and the Global South; conducted evaluations in nine countries; and solicited funds from four of the world's biggest donors.
The Immpact project, Graham explained, sought to enhance the methods for measuring results; design new strategies; and improve communication between researchers and policymakers. This approach, she believes, will improve the evidence base and allow decision-makers to design, implement, and fund more effective safe motherhood strategies. She noted that Immpact addresses the broader strategic level, not individual field programs. Finally, she expressed her regret that there was not enough time to explain the individual tools they created to conduct their assessment.
Unequal Access, Unequal Outcomes
Cynthia Stanton, an assistant professor at the Johns Hopkins Bloomberg School of Public Health and a researcher with Immpact, discussed the many factors that contribute to unequal outcomes in maternal mortality. In-depth analysis of global data confirmed what other studies had shown: a striking relationship between maternal mortality and economic status, both by country and by wealth quintiles within countries. The research also showed, based on 20 countries' Demographic and Health Survey data, a relationship between maternal mortality and the absence of a trained attendant at delivery. Stanton studied a national campaign in Indonesia that proved successful at increasing the percentage of births with an attendant present, but succeeded at increasing the provision of emergency care only to the less-poor quintiles.
Supply and Demand
The next two presentations addressed the supply and demand barriers to care, and the factors that contribute to each. Dr. Julia Hussein, the scientific director of and a senior research fellow with Immpact, began with supply. The title of Hussein's presentation—"Too Few, Too Unskilled, Too Late"—captures her main critiques of childbirth care: the lack of enough birth attendants, a lack of proper knowledge and skills among health providers, and delayed decision-making when proceding to Caesarean birth. In addition, she noted the poor availability of childbirth services (especially to people in remote areas) and inadequate equipment and supplies.
Next, Sophie Witter, a health economist with both Immpact and Ipact, tackled the demand side. Her title was equally telling—"Too Far, Too Costly, Too Unfamiliar." Services are too far away from many population groups, particularly the poor, to develop demand for services. Cost constraints were tremendous, and made worse if Caesarean birth was necessary; incredibly, in some regions, the procedure costs as much as 138 percent of annual per capita income. Low education levels, coupled with low awareness of the existence of or the need for skilled birth attendants, further reduces demand. Gender, cultural, and related factors such as shame and mistrust further complicate interventions. Witter noted that these and other barriers are often interconnected, and that in addition to health concerns, economic outcomes are negatively impacted, as women often need to borrow and/or sell assets to pay for services.
Both Hussein and Witter examined these issues in specific countries in the study, pointing out the complexity of the issues. Hussein noted that addressing cost alone was not enough. In the case of Ghana, where user fees were eliminated, the providers of care were so unskilled that little difference was seen. Similarly, the Indonesian Village Midwife Program increased the numbers of attendants, but failed to improve their skills. And in Burkina Faso, explained Witter, there was a tremendous difference in emergency care between the rich and the poor, the costs of these services could be "catastrophic" for the poor, and traditional taboos influenced decision-making.
The Road Ahead
Graham discussed the future of Immpact, which she described as "closing the loop." The first part of the loop—assessing decision-makers' priorities, and collecting, analyzing, and synthesizing information—has been completed. Immpact spent nearly a year working with country representatives to analyze the goals of decision-makers and several years on data collection, analysis, and the creation of new tools. The next steps—communicating and translating evidence for effective decision-making, and using the evidence to inform program design and implementation—will be critical to ensuring the success of Immpact. Graham notes that, indeed, the emphasis on communication and application of the research is a novel aspect of Immpact. Evidence-based program design "is taken as a given," she said, but that was not true when Immpact began. But closing the loop will be difficult, as there is often a disconnect, both in priorities and language, between researchers and policymakers.
Graham closed by offering several "translated" messages (relating to the mortality-poverty connection, the importance of skilled attendants, and assuring quality of care) for policymakers to apply. She then gave the example of Malaysia, which implemented similar policy changes, with the result that 30 percent of the public expenditure for health went to the poorest quintile of the population. This was a significant reversal of prior policy, and it led to improvements in maternal mortality. Although maternal mortality has been an intractable problem, this example shows that comprehensive research translated into policy-friendly language can make a difference.
By Gib Clarke