Using Web Search Data to Track Dengue Epidemics
Author Information
Author(s): Emily H. Chan, Vikram Sahai, Corrie Conrad, John S. Brownstein
Primary Institution: Children's Hospital Informatics Program, Harvard-MIT Division of Health Sciences and Technology
Hypothesis
Are web search queries a viable data source for the early detection and monitoring of dengue epidemics?
Conclusion
Web search query data can effectively track dengue activity in several countries and provide near real-time information, complementing traditional surveillance methods.
Supporting Evidence
- Models built on Google search volume for dengue-related queries matched official case counts well.
- Web search data can provide near real-time information, unlike traditional reporting which can be delayed.
- The study included data from five countries with significant dengue activity.
Takeaway
People search for information about dengue online, and this study shows that we can use those searches to help detect dengue outbreaks faster than traditional methods.
Methodology
A univariate linear model was built using Google search query data and official dengue case counts from five countries to assess the correlation and fit of the models.
Potential Biases
Potential for panic-induced searching and misdiagnosis of dengue based on search queries.
Limitations
The study is limited by under-reporting of dengue cases and the reliance on internet access, which may not be uniform across regions.
Participant Demographics
The study focused on data from Bolivia, Brazil, India, Indonesia, and Singapore.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website