Jana Howden
Jana Howden completed a double degree in Arts and Science at Monash University in Melbourne, Australia.
A simple Google search could hold the key to tracking and preventing the outbreak and global spread of epidemics.
In research published in PLOS Computational Biology, an international team has used a mathematical modelling tool to track dengue-related Google searches to better predict dengue fever activity.
Dengue fever is one of the fastest-growing mosquito-borne diseases and threatens over half the world’s population. About 390 million people are affected by it each year. The disease can prove fatal, with symptoms ranging from fever and aches to haemorrhage and death.
Despite the huge number of individuals at risk each year, governments rely on hospital-based reporting to monitor the disease, a method hampered by poor communication and time constraints.
In response to the demand for an effective disease surveillance tool, the researchers drew on a simple premise: the more people affected by dengue, the more Google searches on the topic.
“Internet search tracking is consistent, efficient and reflects real-time population trends, giving it strong potential to supplement current epidemiological methods,” the authors explain.
To test their theory, the researchers looked at dengue activity in five countries: Thailand, Singapore, Brazil, Mexico and Taiwan.
By modifying an existing modelling tool – known as ARGO – traditionally used for influenza surveillance, the team tracked the top 10 dengue-related search terms in each geographical region within a given time period.
Government-provided clinical data on dengue incidences was also entered into the ARGO model. The ARGO estimates were then compared to those of five other data tracking models.
The result: in four of the five cases, ARGO gave more accurate results than any of the other methods. Only in the case of Taiwan were the results less accurate, which the researchers speculate was due to inconsistent seasonal disease patterns in the country.
Commenting on the future of the modelling tool, the researchers note that rather than replacing healthcare-based disease surveillance, the tool should help confirm or deny suspected disease trends ahead of traditional surveillance systems.
