All articles Health Tech Business

The surprising role of social media in the fight against pandemics

SARS, Zika, Ebola and now the Wuhan coronavirus, in recent years a series of infectious diseases have hit the headlines. With the world becoming increasingly interconnected, global disease spread has never been easier. Over 100,000 international flights daily allow us to travel anywhere in the world within 24 hours. Which means diseases can too.

Whilst technology may be facilitating the spread of diseases it has also created an unlikely tool in the fight against them - social media. It might seem strange to link twitter feuds to the fight against global pathogens but an array of innovative new technologies are doing just that.

Several studies have shown that information collected from social media during epidemics can be a valuable asset. In contrast to news reported through health institutions and official reporting structures, social media posts are available in real-time and are being shown to be an effective way to detect outbreaks and monitor the spread of diseases.

In one study analysing the twitter posts, Professor Nello Cristianini from the University of Bristol, found that the frequency of tweets containing phrases such as ‘cough night’, ‘sore head’, and ‘swine flu’ aligned with influenza related illness rates throughout the United Kingdom.1

Google Flu Trends 2 was created after it was found that the number of Google searches containing influenza related keywords correlated with the number of people with symptoms. It was shown to predict regional influenza activity 1–2 weeks earlier than the Center for Disease Control and Prevention, however, it was found to consistently overestimate flu prevalence.3

Flu is not the only disease where social media data has been a valid tool. During the 2010 Haitian cholera outbreak, evidence from social media and news articles was used for disease mapping. This information was shown to be available prior to official reports and correlated with official reporting figures.4

The Global Public Health Intelligence Network (GPHIN) uses similar methods and is proving an increasingly useful way of detecting epidemics. Using artificial intelligence (AI) algorithms GPHIN analyses thousands of online sources per day to identify potential disease concerns. These techniques were used for early detection of the SARS outbreak 2002 and hundreds of worldwide disease events since5 including the most recent outbreak of the Wahun coronovirus.

Those who don’t use their phones to tweet, tag or gram may still be contributing to the fight against pandemics. Mobile phone tracking has been shown to be a helpful way of monitoring the movement of people and predicting disease spread. The Harvard School of Public Health has conducted a study in Bangladesh using data from mobile phone networks to track travel patterns with an aim to predict the spread of diseases such as malaria and dengue. 6

As our mobile devices become better at detecting our movements, behaviours and even fluctuations in vital signs it may be possible to detect early signs of illness before symptoms develop. They could be used to relieve pressures on our health services, for example by predicting A&E attendance. With concerns about the negative impact of technology on our lives, we should look to the tremendous benefits social media and AI could have. And in the meantime, just keep tweeting.

Dr Annabelle Painter is a Clinical AI fellow at Babylon Health.



References

1. Lampos V, et al. Nowcasting Events from the Social Web with Statistical Learning. In: ACM Transactions on Intelligent Technology (TIST). Bristol, UK:Department of Computer Science, University of Bristol (Sep 2011) Available: http://www.cs.bris.ac.uk/Publications/pub_master.jsp?id=2001449

2. www.google.org/flutrends/

3. Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L.. Detecting influenza epidemics using search engine query data. Nature 2009; 457:1012-4; PMID:19020500; http://dx.doi.org/10.1038/natu... [PubMed] [CrossRef] [Google Scholar]

4. Chunara R, Andrews JR, Brownstein JS.. Social and news media enable estimation of epidemiological patterns early in the 2010 Haitian cholera outbreak. Am J Trop Med Hyg 2012; 86:39-45; PMID:22232449; http://dx.doi.org/10.4269/ajtm... [PMC free article] [PubMed] [CrossRef] [Google Scholar]

5. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2687026/

6. Mapping imported malaria in Bangladesh using parasite genetic and human mobility data Hsiao-Han Chang et al. eLife. 2019; 8: e43481. Published online 2019 Apr 2. doi: 10.7554/eLife.43481