Harnessing Big Data to Prevent the Spread of Zika
A Global Challenge
Despite many advances in global health and epidemiology, containing viral outbreaks remains a complex challenge — especially with the millions of travellers crossing country borders daily.
Last year, the Zika virus proved to be no exception. Just two years after outbreaks were reported on island nations of the Pacific, the WHO deemed Zika a worldwide epidemic — leading governments scrambling for an effective response for a disease that has a devastating impact on newborn babies.
Given the global nature of epidemics, governments worldwide have begun looking to data scientists for ways to be a step ahead of outbreaks. Answering the call was Simon Hay, director of Geospatial Science at the Institute for Health Metrics and Evaluation (IHME), who led an international team of researchers tasked with converting the multiple types of datasets available into a predictive tool that can help curb the virus’ spread.
Mapping The Spread
The vision of the research team was to use big data to build a detailed map highlighting which locations are susceptible to the spread of Zika, enabling researchers to better understand the change in the geographical distribution of this globally emerging arboviral disease and mobilize smarter responses. With that in mind, both teams dug through a broad base of “Zika” search results from medical publications and extracted all of the location information on past and current outbreaks, which they plotted on a map of the Earth. Next, the team gathered data on a set of six variables they believed to have an effect on global outbreak: suitable viral temperature, vegetation index, humidity, precipitation and urban vs. habitat type. The resulting model allowed the team to generate a map of the globe, in 5 km x 5 km squares, showing how suitable an unaffected area was for the future transmission of Zika.
The findings were first published in an article titled “Mapping global environmental suitability for Zika virus” in eLife and were also presented the findings at a meeting on Zika organized by the U.S. Department of Health and Human Services and the Biomedical Advanced Research and Development Authority (BARDA), where information from the models was used to inform testimony given by BARDA Director Robin Robinson at the U.S. Senate Committee on Health, Education, Labor and Pensions hearing on Zika.
While containing epidemics is still a complex, ever-evolving challenge and much remains to be understood about how to best prevent and control Zika, the map serves as an important first step in predicting the spread of the virus and provides valuable information for policymakers looking to take preventative action. As it stands, these findings have helped inform travel advisories, especially for pregnant women, as well as the allocation of medical and sanitation resources. The project is also an example of how big data can allow governments worldwide to collaborate and disseminate information about infectious diseases much more efficiently. Given how rapidly diseases can spread in this age of international movement and connectivity, keeping citizens, clinicians and governments informed is critical to staying ahead of epidemics.