Internet monitoring to predict infectious disease outbreaks

A review  to be published in The Lancet reveals that for the outbreaks of infectious disease to be accurately predicted, monitoring of search queries in popular search engines may be the best method. Milinovich and colleagues compared data across a number of studies and found that for diseases such as Influenza, H1N1 and Dengue it was possible to predict disease outbreaks weeks in advance using digital data analysis. Analysis of posts on social media is also a useful predictor.


The Office for National Statistics found that 83% of households in the UK have access to the internet, and data in the US suggests that almost half of patients seek medical advice online prior to visiting a doctor and  up to 8 million health-related searches are generated per day in the US. Such a large quantity of data is useful for assessing risk in large populations. Moreover, access to the internet is relatively high in some developing countries compared to access to healthcare. For example, In Ghana 2009 data reveals that there is one doctor per 11,500 people (Oxfam) while one in nine people have access to the internet, meaning that for many the internet is the only source of information. The numbers of Africans having internet access is rising rapidly. Approaches using monitoring of internet search habits within populations may be key to preventing serious disease  in the 21st century

Reference

Milinovich et al.- Internet-based surveillance systems for monitoring emerging infectious diseases The Lancet Infectious Diseases February 2014.