“Do Magic Bullets Kill?” is written at the top of the whiteboard in Dr. Simon Hay’s office at the Institute of Health Metrics and Evaluation (IHME) in Seattle, Washington. It is his mantra for an innovative mapping project that reveals geographic trends and disparities in child growth failure across Africa between 2000 and 2015.
“With these maps, our idea was to think holistically about all the factors that can influence a child’s well-being,” Dr. Hay says. “Were they born to mothers who were well-nourished, in hospitals with skilled health attendants? Were they exclusively breastfed and did they have enough food growing up? I wrote ‘Do Magic Bullets Kill?’ as a reminder not to fixate on just one thing.”
The IHME maps, which published in the March 2018 issue of Nature, are powered by an array of seemingly unrelated data sets: home health surveys, weather reports, agricultural statistics, even light pollution, war zones, and more. They were calculated together using “model-based statistics,” a technique borrowed from the field of geographic information systems (GIS).
“It’s a mathematical way of expressing that two things close together in time and space are more likely to be related than two things located far apart,” Dr. Hay says. “The finished maps tell stories in a way that rows of tables and data can’t.”
And the stories these data illustrate is how far African countries have come in the last 15 years towards meeting the World Health Organization’s (WHO) global nutrition targets for 2025.
“The results alone are astonishing for me – an African accustomed to international headlines depicting a continent consumed by war, famine and hunger,” the late-Kofi Annan wrote in an op-ed in Nature about the publication of IHME’s maps. “The Africa shown in these maps tells a different story: one of measurable, steady progress on issues long thought intractable.”
Annan wasn’t the only person surprised by the on-the-ground reality. When Rahul Rawat, a senior program officer on the nutrition team at the Bill & Melinda Gates Foundation, first saw them he zoomed in on his home country of Zambia. Stunting had decreased from 56.1 percent to 41.6 percent and underweight children had dropped from 21.8 percent to 13.5 percent.
“These maps show levels of progress in Zambia that I didn’t know existed,” Rawat says. The maps provide a critical, deeper layer of information that national statistics lack. “They show that progress has been uneven across the country, with areas of significant progress but also pockets where rates of stunting and wasting have been stagnant over the past 15 years.”
At the current rate of improvement, Zambia will achieve the WHO’s 2025 global target for wasting, which is when someone gradually becomes thinner and weaker. Although, while improvements in stunting are a pleasant surprise, at the current rate of progress it will only achieve a 25 percent reduction by 2025 – far short of the WHO’s target of 40 percent.
The foundation decided to support the IHME mapping project because it gives countries like Zambia a powerful tool to help with the work ahead. Across Africa, the maps can assist decision-making about health policies, programs, and resource allocation. And while the IHME maps are groundbreaking in their use of GIS technology, the data sets that power their algorithms have been collected for decades.
“People think there’s not much data available about Africa, but that’s not true,” Dr. Hay says. “Organizations like UNICEF, the World Bank, and a bunch of health foundations have gathered loads of household surveys. Plus, there are the demographic and health surveys made by the countries themselves.”
The records IHME obtained represent an enormous data set, approximately 58,000 records, including health data for 1.3 million children under age five. Using this information as a base layer, the IHME team added ‘covariant’ data for ancillary factors influencing a child’s nutritional levels, such as exposure to nighttime lights, irrigation, the amount of vegetation, whether there’s urban war, and even elevation.
With the child growth failure map of Africa completed, IHME’s team has moved on to creating more GIS-powered maps for other aspects of global health that can be difficult to visualize. This month, the team published a map for education, and computations are underway for maps about anemia, low birth weight, childhood overweight, and breastfeeding.
The IHME maps may not be a magic bullet, but as Kofi Annan concluded in Nature: “Such fine-grained insight … shows governments, international agencies and donors exactly where to direct resources and support.
“Without good data, we’re flying blind. If you can’t see it, you can’t solve it.”