April 12, 2010
That great reformer Martin Luther said, "If [women] become tired or even die, that does not matter. Let them die in childbirth--that is why they are there." Luther never recanted this statement. I guess he had other fish, or indulgences, to fry.
In the days before modern obstetrical care and antibiotics, maternal death was an equal opportunity killer, killing mothers in all classes of society in about 1% of births. Among the historical victims of maternal mortality were Queen Mumtaz Mahal (during the birth of her 14th child, whose grieving husband built the Taj Mahal as a memorial); Henry VIII's mother and two of his six wives; Mary Wollstonecraft, author of A Vindication of the Rights of Women; Abraham Lincoln's sister; and Theodore Roosevelt's first wife.
Today in the developed world, maternal mortality is mostly a thing of the past (the rate being about 0.01%). This is a reduction of two orders of magnitude since the 1930's. Maternal mortality is, however, not a thing of the past in the developing world, and the United Nations Millennium Development Goals include one of reducing maternal mortality by 75% between 1990 and 2015, along with a goal of reducing child mortality by two–thirds during this same period.
While some conventional wisdom says that a nation needs to develop economically in order to reduce its maternal mortality rate, historical evidence shows otherwise. Poverty is not necessarily the major determinant of a country's maternal mortality rate, though it is of its infant and child mortality rate. In the late 19th century, as improved standards of living were introduced in Europe and America, i.e. better nutrition, hygiene and housing, leading to attendant better health, child and infant mortality rates began to decline and life expectancy increased. Maternal mortality rates, however, did not begin to decline until the 1930's. The risk of women dying in childbirth in 1930 in England was the same as it was in 1850, though the country was far more developed.
Ironically, and the reverse of today, in 19th century England the maternal mortality rate was highest among the wealthy and middle class, who were attended by physicians in hospitals. The lowest rate was among poorer women delivered at home by trained midwives. Doctors often interfered unnecessarily and disastrously in the labor process, including with forceps delivery, anesthesia, and manual removal of the placenta. These interventions too often proved fatal. Trained midwives, who did home deliveries, let nature take its course with better resulting maternal outcomes.
The 1930's brought sulfonamides, which fought childbed fever, as well as blood transfusions, ergometrine and, later, penicillin, as well as better training and organization of obstetrical services. There was also a coincidental reduction in the virulence of the streptococcus virus in the 1930's. All these led to a decline in the maternal mortality rate throughout the Great Depression, indicating that, in general, poverty is not a determinant of maternal mortality, though starvation level poverty may be.
Assuming that the leading causes of maternal death in Europe and America in 1850 are the same as in the developing world today, this gives hope that maternal death can be successfully combated. Since puerperal fever caused about 40% of maternal deaths in 1850 and almost none today in the developed world, this would seem the first priority for attack.
Until this week, it was thought that the maternal mortality rate had not declined substantially in the past twenty years. Maternal mortality was thought to kill over half a million women a year at a rate similar to 1990. The rate in the developing world was thought to be 400 per 100,000 live births today, only marginally less than the rate, 430, in 1990, or about 7% less, and virtually identical to the rate in England in the 19th century. Reducing the rate was proving to be stubborn.
This week The Lancet reported a study from the Institute for Health Metrics and Evaluation (IHME) at the University of Washington and collaborators at the University of Queensland. This study reports that the maternal mortality rate has indeed declined since 1980 from 442 globally to 320 in 1990 and down further to 251 in 1908. The annual deaths are not over 500,000, as previously thought, but 343,000. Until this week, policy makers had been depressed and frustrated at the lack of progress in reducing maternal mortality. This report gives us reason to hope that our interventions are making a difference. The report also pointed out that progress would have been greater but for the HIV epidemic, which leads to about 50,000 of those 343,000 maternal deaths.
Prior research has pointed out that the decline in the rate over the past 75 years has been largely independent of social and economic class. This recent study confirms this. Large declines since 1980 have been revealed in poor countries such as Egypt, Romania, Bangladesh, India and China. Reductions in maternal mortality can happen if obstetrical services are better organized. Previous studies revealed that in one town in northern England in the 1930's, the health community organized better obstetrical care, and within a few years the maternal mortality rate declined from 900 to 170 with no change in the poverty rate. This was contrasted with a prosperous religious community in Indiana, which refused all outside medical care, including obstetrical, with the result that their maternal mortality rate in 1980 was 872, as opposed to 9 in the rest of the state.
The recent Lancet study confirms that when women time, space, and limit their births, their mortality, and that of their children, declines. Prior studies show that about 30% of maternal mortality could be prevented with universal access to family planning. Most of the rest could be prevented with access to trained midwives, modern obstetrical treatment, pre-natal care, blood transfusions and antibiotics. Many deaths could be prevented by access to safe abortion services. The Lancet study says that the reduction in maternal mortality can be attributed to four factors, among others: a reduced birth rate, increased income, increased maternal educational attainment and attendance by skilled birth attendants.
Martin Luther was wrong. Death in childbirth is not women's inexorable fate. Yet, in the United States it is increasingly so. In 1987, there were 6.6 deaths for every 100,000 pregnancies. The number of deaths has climbed to 13.3 per 100,000 in 2006, the last year for which figures are available. Thus, the rate of maternal mortality has doubled in the last twenty years. Lest our northern neighbors get too smug, the rate has also increased in Canada, as well as Norway. The reasons are many, including a different way statistics are recorded, but, according to previous studies, the deaths are mostly among poor and minority mothers. Insufficient access to health care, including pre-natal care, is one factor. Others include the facts that women are increasingly obese and older as they give birth. There are also more Caesarians, which are risky for mothers, a return perhaps to the 19th century experience where medical intervention could do more harm than good.
A comparison of mortality rates between mothers and children is sobering: while 343,000 women die annually from pregnancy-related causes, 8.8 million children under five die annually, approximately twenty-five times more. The rate of child mortality has declined since 1990 in developing countries from 90 to 65 deaths per thousand, (about a 28% reduction) while the maternal mortality rate is only slightly less at 22%. In least developed countries, the child mortality rate is 130 per 1,000 annually. Poverty is the major determinant of child mortality, with the proximate causes being lack of access to safe water, malaria, pneumonia, other infections, lack of oxygen at birth, and preterm delivery. Vaccines, antibiotics and other low cost interventions would prevent about half of child deaths. There is some intersection between interventions to reduce maternal deaths and those to prevent infant deaths, like pre-natal care, improved obstetrical care and antibiotics, but other interventions for the infant are separate and dependent on a public health care and economic system being functional, especially with vaccines, safe water and mosquito netting to prevent malaria.
In historical times, the difference in annual deaths between mothers and children was greater. It is estimated that the child mortality rate in ancient times was about 50%. In other words, one half of babies born did not survive to adulthood. Estimates range from child mortality rates of 50% in Rome in 200BC-200AD; 48% in Japan in 1300-1400; 50% in France and Sweden in 1600-1700; and modern hunter-gatherer tribes of 46%.
There was an extraordinary difference between the evolutionary mortality rates between mothers and children ⎯ 1% for mothers and 50% for children. Modern medicine, carefully applied, along with fewer births, well spaced and timed, have reduced these awful death tolls. The millennium development goals of reducing these rates further are laudatory. It shows that the world cares and is watching. This is not the time to be disheartened. Progress has been made in our time as well as in the time of our grandparents. I hope our grandchildren will be able to say the same.