In the Journals

Search engines effective in predicting dengue prevalence

The use of a model that tracks internet search queries may be effective in anticipating the increased spread of dengue and other neglected tropical diseases in countries that are less developed, a potential benefit for hospitals and travelers alike, according to a study published in PLOS Computational Biology.

“The global spread of the internet has opened up the opportunity to investigate whether users’ activity patterns on internet search engines and social media platforms may lead to reasonable estimates of dengue infection levels,” Shihao Yang, a PhD student in the department of statistics at Harvard University, and colleagues wrote. “In theory, internet search tracking is consistent, efficient and reflects real-time population trends, giving it strong potential to supplement current epidemiological methods.”

To assess the efficacy of an AutoRegressive model with Google search queries as exogenous variables (ARGO) in tracking dengue in five countries where the disease poses a local threat, the researchers examined historical dengue incidence from government-led health care agencies in Mexico, Brazil, Thailand, Singapore and Taiwan. The model was previously implemented to track influenza using influenza-related search queries.

Additionally, the researchers also used dengue-related search results from Google, which were collected at the national level. All data were collected between 2001 to 2016, with varying time frames listed for each country.

In Brazil, Mexico, Thailand and Singapore, all benchmark models across nearly every accuracy metric were outperformed by ARGO, with decreased errors observed in relation to peaks, off-season periods and periods with low levels of infection. Overshooting was resolved using the program, especially in Mexico between 2006 and 2008 as well as 2012 through 2014. Improvements in overshooting were also observed in Brazil between 2006 and 2010.

However, the results observed in Taiwan were remarkably different, with ARGO not outperforming benchmarks in this location. The researchers speculate that this may be because little prevalence was observed, with the exception of two spikes in 2014 and 2015 as opposed to the seasonal fluctuation observed in the other four countries.

“The wide availability of [the] 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,” Mauricio Santillana, PhD, from Boston Children’s Hospital and Harvard Medical School, said in a press release. “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.” — by Katherine Bortz

Disclosure: The researchers provide no relevant financial disclosures.

The use of a model that tracks internet search queries may be effective in anticipating the increased spread of dengue and other neglected tropical diseases in countries that are less developed, a potential benefit for hospitals and travelers alike, according to a study published in PLOS Computational Biology.

“The global spread of the internet has opened up the opportunity to investigate whether users’ activity patterns on internet search engines and social media platforms may lead to reasonable estimates of dengue infection levels,” Shihao Yang, a PhD student in the department of statistics at Harvard University, and colleagues wrote. “In theory, internet search tracking is consistent, efficient and reflects real-time population trends, giving it strong potential to supplement current epidemiological methods.”

To assess the efficacy of an AutoRegressive model with Google search queries as exogenous variables (ARGO) in tracking dengue in five countries where the disease poses a local threat, the researchers examined historical dengue incidence from government-led health care agencies in Mexico, Brazil, Thailand, Singapore and Taiwan. The model was previously implemented to track influenza using influenza-related search queries.

Additionally, the researchers also used dengue-related search results from Google, which were collected at the national level. All data were collected between 2001 to 2016, with varying time frames listed for each country.

In Brazil, Mexico, Thailand and Singapore, all benchmark models across nearly every accuracy metric were outperformed by ARGO, with decreased errors observed in relation to peaks, off-season periods and periods with low levels of infection. Overshooting was resolved using the program, especially in Mexico between 2006 and 2008 as well as 2012 through 2014. Improvements in overshooting were also observed in Brazil between 2006 and 2010.

However, the results observed in Taiwan were remarkably different, with ARGO not outperforming benchmarks in this location. The researchers speculate that this may be because little prevalence was observed, with the exception of two spikes in 2014 and 2015 as opposed to the seasonal fluctuation observed in the other four countries.

“The wide availability of [the] 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,” Mauricio Santillana, PhD, from Boston Children’s Hospital and Harvard Medical School, said in a press release. “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.” — by Katherine Bortz

Disclosure: The researchers provide no relevant financial disclosures.