Would you rather give birth in 1900 or 2018?

The answer to the question ‘would you rather give birth in 1900 or 2018’ might depend on where in the world you are. In 1900 in Sweden, before the construction of modern hospitals or the use of modern medicine, 228 women died per 100,000 live births; in Sierra Leone today, that number is estimated at 1,360 per 100,000, the highest in the world (compared to around 50–100 in most other countries, though Sierra Leone is not a distant outlier among low-income countries). With the Sustainable Development Goals (SDG) setting out a target of maternal mortality rates of less than 70 per 100,000 live births, this International Women’s Day is a good time to ask what Sweden did right, and how can policymakers learn from it? The answer is both simple and complex…

While great strides have been made towards achieving this target in Sierra Leone and elsewhere — the President of Sierra Leone introduced the Free Health Care Initiative (FHCI) in 2010, abolishing health fees for pregnant women, lactating mothers, and children under five — more needs to be done to complement these positive steps.

What can we learn from Sweden?

Sweden’s success story offers an early model of comprehensive data collection and effective statistical analysis. The story can be traced back at least as far as the early 1600s. The Swedish Clergy began developing an extensive information system — and over the next 100 years a number of new indicators were added (including births and deaths), capturing almost the entire population. An Office of the Registrar General was founded to compile and analyse the numbers, and in 1751 the first national statistics on maternal mortality were presented.

These revealed that around 900 women had died per 100,000 pregnancies. What really caught the government’s attention, however, was when the Commission for Health argued that of 651 deaths, 400 could have been saved if there had been more midwives. The key observation was not that a lot of women were dying, but that most of the deaths were avoidable.

The government’s response? Training and deploying more midwives, aiming for a basic coverage of a midwife per parish. By 1860, 40% of deliveries were attended by a trained midwife and mortality rates had fallen to nearly half their previous figure. By 1900, 150 years after the harrowing level of avoidable deaths was first noticed, close to 80% of deliveries were attended by a trained midwife, and the rate of 228 deaths per 100,000 was the lowest in the world. To put this into context, the rate in England and Wales at the time was 450 and 900 in the USA.

It wasn’t until the 1930s that confidential investigations into maternal deaths started in the USA; a mission to Sweden helped the US government discover that a policy on professionalising midwifery would be the answer — despite earlier opposition from doctors. Once data was available, action could be taken — and, from the late 1930s, trajectories in Sweden, the USA, and England and Wales converged. Continuous medical improvements in medical technologies have brought the mortality rate to below five per 100,000 pregnancies today.

The importance of data and responding to it

In order to facilitate an effective system, the collection of data must not be an afterthought or a disconnected evaluation, but an instrumental component in bringing about positive change. This can be difficult in low-income countries that often lack the vital registration systems and institutions to accurately record births, deaths and the causes of death. Yet, what the Swedish example shows us is that this stage is often essential for effecting change — after all, we can’t manage what we can’t measure.

By recording not only the number of maternal deaths but also the cause of death, to highlight the extent of ‘avoidable deaths’, we can begin to think about the next stage: appropriate interventions. An important example can be found in 1930s America, where it was these data that began challenging the still-held assumption (from both doctors and the wider public) that high levels of maternal mortality are somehow inevitable in certain places. Before this point, midwives in America had not only been viewed as outsiders by the medical profession, they had been blamed for maternal deaths (which, in fact, it has been argued that doctors’ over-willingness to use unnecessary procedures had endangered women).

Navigating complex systems

Only when this sort of assumption is tackled head-on, and tested with accurate data, can countries move onto the next stage of identifying measures to improve maternal healthcare. While there are many possible interventions — from improving hand hygiene to abolishing user fees — it’s important to remember that the complex nature of health systems means no intervention is likely to be effective in isolation.

A health system brings together many interconnected elements: service delivery policies, human resources for health policies, education policies and more, encompassing many stakeholders and issues ranging from equitable access to healthcare to women’s social status. Professionalising midwifery is not a panacea on its own, without an appreciation women’s rights and cultural attitudes to birth and economic status, but we must get the basics right while simultaneously recognising the big picture.

History teaches us that it is possible to make significant improvements in maternal mortality without expensive hospital technologies. A well-developed information system that records the cause of death enables governments to identify the life-saving strategies best suited to them within the context of a wider health system. With hundreds of millions of dollars in government and donor money currently flowing into health sectors in low-income countries, and the ambitious goal set out in the SDGs, we need to remember not to overlook the basics of detailed information and straightforward hygiene as building blocks of resilient health systems that support improved maternal health.