- by Gary Finnegan
Tracking dengue: how Google searchers can help
A new study by researchers at Harvard and Break Dengue shows the power of real-time disease monitoring. It could even enable swifter responses to disease outbreaks
A new research paper published in PLOS Computational Biology, a peer-reviewed scientific journal, reveals how search engines can be used to monitor dengue fever outbreaks in underdeveloped countries.
The paper is authored by scientists at Harvard Medical School, Harvard University, Boston Children’s Hospital and Nicholas Brooke, Executive Director of The Synergist which runs Break Dengue.
It looks at how big data analytics combing Google searchers with official clinical data from government sources can give a quick and accurate picture of dengue outbreaks. The dengue tool builds on previous work by the Harvard team which uses an approach to mathematical modeling known as AutoRegression with Google Search queries (ARGO).
While dengue infects around 390 million people every year, accurate monitoring has proven challenging. ARGO has revived hopes that search engine data could help officials to track disease outbreaks. Previous efforts, such as Google Flu Trends and Google Dengue Trends had not delivered the level of accuracy that had initially been expected.
The new study used a modified version of ARGO to study dengue fever activity in Brazil, Singapore, Thailand, Taiwan, and Mexico. Researchers tracked the most frequently searched dengue-related search terms in each country and collected history dengue data logged by official government health authorities. Both datasets were then put into ARGO which calculated a “near real-time” dengue prevalence value for each country.
By comparing the ARGO estimates to data from five other surveillance tools, it was concluded that ARGO gave a more accurate picture than any other approach for Mexico, Singapore, Thailand, and Brazil. The results for Taiwan were less accurate but the authors attribute this to Taiwan’s less consistent patterns of dengue outbreaks from one year to the next.
The study shows the potential of search engines such as Google to deliver accurate and near real-time data on mosquito-borne diseases in countries where surveillance systems may be weak. The group is now looking at how the dataset could be further bolstered by incorporating information from other sources – such as weather and environmental data.
‘The wide availability of internet throughout the globe provides the potential for an alternative way to reliably track infectious diseases, such as dengue, faster than traditional clinical-based systems,’ says study senior author Mauricio Santillana of Boston Children’s Hospital and Harvard Medical School. ‘This alternative way of tracking disease could be used to alert governments and hospitals when elevated dengue incidence is anticipated, and provide safety information for travelers.’
Adding crowd-sourced data into the mix is also being explored as scientists aim to build a people-powered surveillance tool that could become an invaluable alert system for health authorities and citizens.
The Harvard team is implementing to building blocks that could shape such a platform. One is the Healthmap.org/denguetrends website. The other is Break Dengue’s Dengue Track initiative. This tool offers a user-friendly online chat system to map dengue fever cases. In return, members of the public who contribute their data are given access to toolkits that help to reduce their risk of infection.
The researchers note that the same modeling system could be used to map other infectious and mosquito-borne diseases – including Zika and malaria. Perhaps you and I are the future of disease surveillance.
Perhaps you and I are the future of disease surveillance.
Click below to begin reporting dengue activity near you to Dengue Track.