Some 350,000 women die each year during pregnancy or soon after giving birth, with women in sub-Saharan Africa, Pakistan and Afghanistan facing the highest risks. Although global rates of maternal death have been dropping by about 1.5 percent each year since 1980, there is still a long way to go if countries hope to meet United Nations Millennium Development Goal (MDG) 5 by 2015—a 75 percent reduction in the number of maternal deaths per 100,000 live births from 1990 levels. Today, an average of 251 women die per 100,000 births, and only 23 countries are on track to reach the MDG, with some countries even moving in the wrong direction.

Approximately 15 percent of women everywhere develop pregnancy complications. These include infections (such as HIV), pregnancy-induced hypertension, obstructed labor and hemorrhage. Most can be managed with proper care, but many women simply do not receive it. "It's not that you require a scientific breakthrough to solve these situations—in developed countries the knowledge and technical tools have been available to women for many, many years," says Ana Langer, coordinator of the Dean's Special Initiative on Women and Health at the Harvard School of Public Health. In many places, however, "women do not have access to them."

In Africa, for instance, women often have to rely on the willingness of others to get obstetric care because of their low social status. "For women to have timely access to services that can help to save her life she still needs the community, husband and family to be able to provide access and transportation," explains Grace Kodindo, an assistant clinical professor of population and family health at Columbia University's Mailman School of Public Health.

Possibly as a result of these societal barriers, since 1980 the Ivory Coast and Zimbabwe have both experienced increases in maternal mortality rates—in Zimbabwe deaths have jumped by 5.5 percent each year in the past two decades, in part because of low female social status and ongoing political conflict. But even the wealthiest countries face problems in maternal health, including the U.S., where 17 women die per 100,000 live births. Inequalities, however, still leave the poorest and least educated women at the greatest risk. "The context may be different, but the bottom line is the same—they don't have timely access to care," Langer says.

Whereas maternal care seems to generally be improving—albeit slowly—statistics on women who die during pregnancy are notoriously fuzzy. Record-keeping is often poor, in part because many women die in their homes rather than in clinics—and even when deaths are noted pregnancy status is not always referenced. According to Kodindo, health care workers are sometimes afraid to report maternal deaths because they believe the information will be held against them. Governments need to "inform the community that the data will be used to improve the health system," she says, not to assign blame. In addition, some deaths that occur during pregnancy are not attributable to it, although records do not always reflect the distinction. Ultimately, it is unclear just how much trust researchers should put in the numbers when assessing progress in improving maternal health.

Researchers are confident, however, about which interventions help the most. Assisted delivery comes out on top because most pregnancy deaths, irrespective of region, occur during labor. "There are still many parts of the world where the mode of delivery is the home, and it's difficult to really deal with emergencies in that setting," says Robert Black, a professor of international health at the Johns Hopkins Bloomberg School of Public Health. If women cannot get to a clinic, they should be accompanied at home by a skilled attendant, he notes.

Another part of the goal is to ensure that women everywhere have access to family planning services, as this reduces the number of unplanned and potentially complicated pregnancies—especially in adolescent and older women. "If family planning reduces the high-risk, high-parity women who have had many children and are at higher risk of death, then certainly it should reduce the maternal mortality ratio," Black says. Such services would also prevent deaths associated with unsafe abortions, which, according to a 2009 editorial published in The Lancet, kill eight women around the world every hour.

Ultimately, three quarters of gestation-related complications are treatable, Kodindo says, but the risks linger in part because many governments have been slow to prioritize maternal health. The number of maternal deaths per 100,000 live births simply has not fallen by the targeted 5.5 percent each year, so most low-income countries are unlikely to reach the MDG target. But experts remain cautiously optimistic. "I hope that in 2015 there will be an assessment of progress and a commitment to make further progress," Black says. "After all, the 2015 targets, even if achieved, are not by any means the lowest possible mortality levels. Substantial effort will still be needed."