Chuck Bednar for redOrbit.com – Your Universe Online
Wikipedia page views could help predict potential disease outbreaks weeks before official health advisories are issued, researchers from Los Alamos National Laboratory report in a recently-published study.
Their research, which was published Thursday in the journal PLOS Computational Biology, indicates that they were able to forecast flu and tuberculosis outbreaks four weeks in advance by monitoring articles on the collaboratively-edited online encyclopedia.
Dr. Sara Del Valle and her Los Alamos colleagues said they were able to successfully monitor outbreaks of influenza in the US, Poland, Japan and Thailand, dengue fever in Brazil and Thailand and tuberculosis in China and Thailand. They also said they were able to forecast all but one of those outbreaks at least 28 days in advance.
Their findings suggest that people have the habit of searching websites such as Wikipedia for disease-related information before actually seeking medical attention, and shows the potential for “training” computer models using public health data in one location and then implementing it in another part of the world.
“A global disease-forecasting system will improve the way we respond to epidemics,” Del Valle said in a statement. “In the same way we check the weather each morning, individuals and public health officials can monitor disease incidence and plan for the future based on today’s forecast.”
According to BBC News, Del Valle and her colleagues tracked the page views of disease-related Wikipedia pages from 2010 and 2013. They tracked the languages that the information on those pages was written in, using that as a way to approximate where those individuals lived.
The data was then compared to actual disease outbreak information provided to the research team by various national health surveillance officials. In eight out of 14 cases, the British news organization said that there was a clear increase in page views in the four week period before health officials declared an outbreak.
Furthermore, the model was able to predict every outbreak except the tuberculosis one in China, but as Del Valle explained, “the goal of this research is to build an operational disease monitoring and forecasting system with open data and open source code,” and that their new study “shows we can achieve that goal.”
Others, however, are not so sure. Dr. Heidi Larson of the London School of Hygiene and Tropical Medicine told BBC that the Los Alamos team’s findings were “compelling,” but that she would still be “wary” about using this method as a tool for predicting outbreaks of all types of diseases.
Dr. Larson explained that, in order for this technique to be effective, the individual would have to have Internet access, be literate, understand how Wikipedia works and have knowledge about the condition itself. She also said that other factors could drive people to check Wikipedia, such as a research paper.
In addition, while online trends could provide valuable signals about disease, Dr. Larson said there were still doubts that the data could be used in policy-making or for intervention, especially in some parts of the world. “I’m not sure how much Wikipedia is used in Africa,” she said. “For issues like Ebola, I don’t think people at the beginning of the outbreak in West Africa would have [been searching], because they wouldn’t have had it [Ebola] before.”