“What is the most dangerous animal in the world?” asks Mr. Engø-Monsen, senior researcher at Telenor Research. “You might answer a polar bear, wolf, lion or shark. But it’s the mosquito.”
The death toll caused by mosquito-borne diseases climbed to 830,000 people globally last year. The second biggest killer? People, who murdered 580,000 of their fellow human beings–a third less than the carnage caused by mosquitoes.
In Thailand, Thai people infected dengue virus shot up to 10,000 people and 27 people died from the infection. Malaria, too, is endemic to our region. And a new strain that resists conventional treatments is now spreading from Cambodia and into Vietnam, with a few cases already making their way to Thailand.
What if we could prevent the spread of these diseases by using big data to predict the geographic spread and timing of the epidemics? Accurate predictive models identifying changing vulnerability to the outbreaks are necessary for epidemic preparedness and containment of the virus.
In particular, accurate predictions will help health organizations to distribute mosquito nets, provide mobile clinics and raise awareness in the affected areas.
But how to trace mosquito movements when their size is too tiny to install GPS tracking devices on them? Fortunately, mosquitoes spend their lives within a 50-100 meter radius. It is humans who spread mosquito-borne diseases over long distances. An infected human may, for example, travel from Chiang Mai to Bangkok where they will be stung by a local mosquito. Once that mosquito is infected, it will infect other people in Bangkok.
Fortunately, almost 100% of Thais own a mobile phone, meaning mobile phone data offers an accurate measures of human aggregation and movement, representing a unique source of information on the human determinants of geographic expansion of emerging epidemic diseases.
During 2013 to 2015, Telenor Research in partnership with the Harvard T.H.Chan School of Public Health, Oxford University, the US Center of Disease Control and the University of Peshawar in Pakistan, demonstrated the power of mobile data to predict and track the spread of epidemic disease.
Mr. Engo-Monsen gathered an estimate of travel path that was extracted from Telenor Pakistan’s data with 40 million subscribers. Subsequently, the actual dengue cases that were reported between 2011 and 2013 were mapped with the travel paths.
The resulting model allowed researchers to estimate the seasonal dengue virus importation and generate dynamic risk maps with direct applications to dengue containment and epidemic preparedness.
All in all, the study showed the potential for call records to accurately reveal mobility patterns that can help combat and predict the spread of virulent diseases.
Presently, Mr. Engo-Monsen, on behalf of Telenor Research, is working with dtac and the Bangkok-based Mahidol University, Harvard T.H.Chan School of Public Health and the Thai Ministry of Health to study on the same topic, but applied to Thailand.
“We hope that the study could be extended to cover the whole region in the future, as Southeast Asian countries have long been afflicted with mosquito-borne viruses,” said Mr. Engo-Monsen.