How Zambia’s partnership with software companies can accelerate elimination efforts
By Stacey Naggiar
Advocacy and Communications Officer, MACEPA, and Global Health Corps Fellow
Allan Walker is building a map. A career computer geek, he’s comfortable using sophisticated software to unveil patterns among massive amounts of data that can be used to help solve some of the world’s most complex problems.
“I’m actually tracing structures right now, as I’m talking to you,” Walker told me over the phone from his home in San Francisco. The structures he’s outlining are buildings in Livingstone, a district in the Southern Province of Zambia with roughly the same area as New York City. But the map he’s developing isn’t just any map. This map will be used as part of a nationwide effort to make Zambia malaria-free by 2020.
To become one of the volunteer mappers, first sign-up for a free account on OpenStreet Map. Once you have created an account, join the Zambia Mapping Project and follow the instructions to get started. You can get involved by charting new parts of the district and validating areas others have worked on.
Success against malaria breeds new challenges
Malaria, a mosquito-borne illness, has seen remarkable reductions worldwide with mortality rates declining 60 percent since the year 2000. But this preventable disease is still killing too many people in sub-Saharan Africa; in 2015, the region had 88 percent of global malaria cases. In Zambia alone, it’s estimated that nearly 3,000 people die from malaria each year.
Since 2005, Zambia’s Ministry of Health, with support from the Malaria Control and Elimination Partnership in Africa (MACEPA), a program at PATH, has been aggressively fighting the disease. Through a combination of interventions such as insecticide-treated bed nets, indoor residual spraying, and now mass drug administration—where entire communities are treated with antimalarial drugs—Zambia’s Southern Province has seen a 93% reduction in malaria in children between 2012 and 2015.
As Zambia sees fewer and fewer cases of malaria, it is of utmost importance that community health workers—think of them as an army of malaria volunteer firefighters—quickly track and respond to new cases. Now imagine the community health workers with a crystal ball that could predict when and where new cases would appear. That would make the job a lot easier, right?
Enter mapping and data visualization
“We want to know, in low incidence areas, where flare-ups might occur so Zambian health officials can mobilize appropriate resources and address them,” says Jeff Bernson, PATH’s director of Results Management, Measurement, and Learning.
Mapping as a public health tool is not new. “Since the dawn of epidemiology in the 1850s we have been combining data and cartography to better understand disease and transmission,” says Bernson. “The difference is now we’re using new, faster, easier to use, and more accurate geo-spatial tools to help us. ”
In 1854, John Snow famously mapped the city of London during a cholera outbreak. But in the middle of the 19th century people had to go door to door to build a map. Today, volunteers like San Francisco-based Walker are mapping from the comfort of their own homes. And anyone can do it.
“It’s fun and quite easy to do. My mother who’s approaching 70 has traced buildings,” Walker says.
The crystal ball: tracing to save lives
In an initiative to #VisualizeNoMalaria, Mapbox, along with the Tableau Foundation, Alteryx, and other partners, is partnering with PATH to support community health workers and Zambia’s Ministry of Health in their quest to eliminate malaria. Mapbox, a software company that crowdsources open-source mapping data, provides the geographical information to Tableau with support from Alteryx’s data blending and analytic software. Tableau’s platform then enhances how the data are used and visualized to solve problems.
How exactly does it work? Volunteers start by logging into the tracing software, HOT OSM (Humanitarian Open Street Map) and examining aerial photographs, drawing outlines around each building. They fill in the shape and then the building is assigned a value.
That value is then merged with other geographical factors—elevation, population density, topographical wetness (where water collects), rainfall, and temperature data—to generate a figure for how likely malaria is to occur in different seasons.
With all this data, algorithms behind the scene generate a score that predicts where malaria cases will pop up. It’s called a “predictive model,” and it’s the kind of crystal ball that community health workers can use to track and treat incidents faster, meaning new cases can be prevented and lives can be saved. It also paves the way for improved decision-making at the Ministry level, ensuring that products, such as bednets, reach the right places at the right times and that expensive medications don’t expire on the shelf.
Since the project to map Livingstone began in mid-July, more than 50 volunteers have mapped 89 percent of the district. That’s some 27,000 buildings in just a couple of months. Now, the race is on to finish the job.
“It’s about the ability to look at a problem holistically and visually. Even with a large-scale problem, you can visually simplify the patterns and really see and understand your data,” Walker explained. “What’s more,” he said, “It’s an opportunity to do work that’s impactful as a volunteer.”
To learn more about how PATH, Tableau, Mapbox, and others are working with Zambia to eliminate malaria, visit www.visualizenomalaria.org