Measurement Matters – The decline of maternal mortality

OWID presents work from many different people and organizations. When citing this entry, please also cite the original data source. This entry can be cited as:

Max Roser (2017) – ‘Measurement Matters – The decline of maternal mortality’. Published online at Retrieved from: [Online Resource]
Our World in Data presents the empirical evidence on global development in entries dedicated to specific topics.
This short post draws on data and research discussed in our entry on Maternal Mortality.

What could be more tragic than a mother losing her life in the moment that she is giving life to her newborn? The visualization below shows how common this once was. In Finland—the country that today boasts the lowest maternal mortality in the world—we see that during much of the 19th century, maternal mortality hung between 800 and 1,000 deaths for every 100,000 births. In other words, each birth came with up to a 0.9% chance of death. And since women in Finland gave birth to 5 children on average at the time, this meant that the death of a new mother was no rare occurrence.

The visualization also shows how maternal mortality has changed since then. Starting in the second half of the 19th century, the risk of dying during or after pregnancy has declined from 0.9% to 0.003%. Giving birth today is 300-times safer than just a few generations ago. How was this possible?

# Measurement and progress: The story of Ignaz Semmelweis

The story of the decline of maternal mortality is a story about scientific research and the importance of measurement. It began with an accidental controlled trial.

In the mid-19th century, the largest maternity hospital in the world was in Vienna, and the most common cause of maternal deaths was puerperal sepsis (an infection of the uterus that killed new mothers shortly after childbirth).1 There was an unexplained but stark pattern taking place in the Vienna Maternity Hospital. The institution was divided into two wings: doctors and medical students were taught in the first, and midwives were taught in the second. In the first wing, every tenth (!) mother died of puerperal sepsis, while in the second ‘only’ one in 25 died. Crucially, women were admitted to the clinics not on the basis of how complicated their case was, but simply on alternate days—in other words, by random allocation. Unintentionally, the hospital had created the conditions suitable for a natural experiment.

In 1846, the young Viennese doctor Ignaz Semmelweis was appointed to the hospital. And it was his suspicion that differences in the routine practices of the two wings would give him insight into what was causing the alarmingly high mortality rates in one. He noticed that in the first wing, where 10% of mothers perished, medical students alternated between helping women to give birth and attending post-mortem examinations of women who had died of puerperal sepsis. They carried out their work in ordinary daytime clothing rather than clean, white coats, and between autopsies and deliveries, they did not wash their hands. By contrast, the pupil midwives in the second clinic did not perform postmortem examinations. And thus, many years before the role of bacteria in diseases was known, Semmelweis discovered that the disease under his study was contagious, and that doctors were transmitting it.

Semmelweis also found that a chemical would destroy the agent that led to the disease, and therefore insisted that his students disinfect their hands with it before attending the labour wards. This led to dramatic results. By 1848, the maternal mortality rates of the two clinics in the Vienna Maternity Hospital were comparable.2

However, Semmelweis’s insistence on his discovery that puerperal sepsis was contagious did not fit into the accepted medical understanding of the time, and this meant an agonizingly slow process of convincing the medical establishment of the importance of better hygiene and measures of antisepsis. Only in the second half of the 19th century was antisepsis widely introduced into routine obstetric practice. But this change, coupled with the discovery of antibiotics and the development of blood transfusion in the first half of the 20th century, sharply drove down the risk of dying in childbirth.

Semmelweis’s story tells us two things. First, it shows how powerful measurement can be. Decades before the modern germ theory of disease was accepted, and with no understanding of the mechanism that caused puerperal sepsis, it was the measurement of the distribution of maternal deaths that gave Semmelweis the crucial insight about what steps could be taken to effectively reduce maternal mortality. Second, it tells us how frustratingly long it can take until new discoveries become accepted and finally change practices. Semmelweis’s discovery, which could have led to immediate and widespread improvements, was not widely circulated. More time would have to pass, with more knowledge accumulated, before steep reductions in maternal mortality could take place.

# Maternal mortality today — and where we could be with best practices

In the chart at the beginning of this article, the short purple line tells us that in 2015 the risk of dying in childbirth stood at 0.216% per birth. Currently, the world sees around 140 million births per year. At the rate of 0.216% this means that globally, we expect 303,000 mothers to die this year.

Imagine what it means for loved ones to lose a new mother. Now try to imagine the depth of this devastation 303,000 times over. This is the reality of 2015. The sloping purple line in the chart tells a story of immense improvement, but it also tells a story that still involves immense tragedy.

The world map below shows the distribution of the risk associated with childbirth today. The global average discussed above hides huge differences in global health. The risk of maternal death is higher than a startling 0.5% in several central African countries.

It is still a very long way down for the countries with the poorest health today. But there is something important and positive to be gleaned from the historical perspective seen here. Finland, the country with the lowest maternal mortality rate today, had a higher mortality rate than today’s worst off countries only a few generations ago. Globally, today’s maternal mortality rate is similar to that seen in the US in the years after the Second World War. There is no reason to believe that what was possible for Finland and the US should not be possible elsewhere.

We have the advantage of well-established medical and technological knowledge and this means that today we are capable of reducing maternal mortality at a much faster pace. You can add Tanzania or Cambodia to the first chart above to see how very rapidly some countries have reduced maternal mortality recently.

Given this foundation of knowledge, what can we achieve in the world as a whole? One way to think about this question is to ask how many lives would be saved if the maternal mortality rate of Finland applied to the whole world. The 303,000 annual maternal deaths would be reduced to 4,217. A unique kind of devastation would be avoided nearly 300,000 times.

To a Finnish woman living 150 years ago, today’s vastly improved reality would have likely been unimaginable. But a reduction to 0.003% is not unimaginable, it is what we know is possible. Our history and experience should encourage us to believe that much better global health is not only possible, it is also achievable if we continue to rely on good measurement and apply the best scientific understanding of what is causing the deaths of mothers.