A year ago, in January 2020, many internet users shared their unusual dry cough symptoms on social media, long before it was discovered that they were suffering from a new form of coronavirus. This is what makes Italian economists and statisticians say that social networks could be a good tool for epidemiological surveillance. “They could help intercept the first signs of an epidemic, before it proliferates undetected, and also monitor its spread” underlines Massimo Riccaboni from the IMT School for Advanced Studies in Lucca, Tuscany.
To support his statements, the economist and his colleagues from the University of Calabria, have analyzed Twitter data to uncover early warning signals of COVID-19 outbreaks in Europe during the 2019-2020 winter season, before the first public announcements were made. Among the data of more than 570,000 unique users and more than 890,000 tweets, they searched for tweets from European countries with the keyword “pneumonia” (in the 7 most spoken languages, namely English, German, French, Italian, Spanish, Polish and Dutch) posted last winter. They then went back to 2014. Then they did the same with other symptoms like “dry cough”.
Tweets two weeks before the first cases of contagion
The results of their study have just been published in the journal Nature. For Italy, tweets showed signs of the first clusters during the first week of 2020, a few weeks before the official announcement of the first case (February 20, 2020). A similar pattern was observed in France. While in Spain, Poland and the UK, signals of the disease appeared 2 weeks before their official cases.
“Messages about symptoms related to COVID-19 predated official public announcements about local outbreaks and were concentrated in areas that would later become hotspots of infection,” the researchers explain.
However, they warn that the research cannot be used for diseases that are still unknown. “Searching for the word ‘pneumonia’ would have been useless before pneumonia is publicly linked to COVID-19″ they insist. But following social networks could prove useful for identify points of resurgent infections and help counter the threat of recurrent pandemic waves.
Source:
- Early warnings of COVID-19 outbreaks across Europe from social media, Scientific reports, January 25, 2021
Read also :
- Loss of smell and taste : the testimony of Lauren infected with the coronavirus
- Screening: rapid PCR test, blood test… for whom
- If 70% of the population wore a mask, the Covid-19 would be stopped
