Tweeting out the news that you just vomited might not be the sexiest thing you could do, but it may very well be the most helpful. That’s the recent plea made by the UK Food Standards Agency, which is encouraging Brits to tweet when they vomit in order to track the United Kingdom’s levels of norovirus.
Norovirus is a very contagious virus that causes inflammation in the stomach and intestines, typically leading to stomach pain, nausea, diarrhea, and vomiting. It’s a nasty affair that begins when people somehow get the poop or vomit from infected people into their mouths. This typically happens when people share utensils with someone who is infected, eat food prepared by someone with norovirus, or touch surfaces or objects that have been contaminated, then put their fingers in their mouths. Gross.
When experimenting with various ways to find data, the FSA determined that social media was the best way to collect them in real time. The BBC reports, when the FSA compared sample data from infected people to relevant tweets, it found a “really good” correlation between the incidence of illness and references to being sick on Twitter. With the use of a model that searches for tweets using relevant words, they can accurately predict an increase in norovirus 70 to 80 percent of the time.
The FSA is issuing a request now for people to tweet if they think they have norovirus so that they can roll out an intervention before a larger outbreak. It tells the BBC that they typically have three weeks before a predicted increase becomes a national concern. If in the impending weeks they do predict a national outbreak, they are going to run a digital campaign on the best practices for dealing with norovirus.
Unfortunately, best practices don’t include any specific medicine. Because the infection is viral, not bacterial, it cannot be treated with antibiotics. The best thing is to rest and to make sure you are hydrated.
While the use of social media to track illnesses has been a popular point of study for academics over the past five years, it hasn’t been readily accepted by national health agencies. While the United States Centers for Disease Control and Prevention does collaborate with Google to analyze the data of flu-related web searches and search for signs of outbreak, it predominantly relies on outpatient reporting and virological test results from laboratories.
In January 2016, the CDC launched a website called “FluSight: Seasonal Influenza Forecasting” that highlights the work of different private research teams that use different forecasting models for flu activity. Since 2013, the CDC has hosted a contest in which teams are encouraged to use data from social media and internet search engines to predict flu activity.
In a statement in January, the CDC said that it uses this competition to “channel private sector resources into the search for innovative approaches to flu forecasting because this approach was more cost-effective than relying on traditional funding and governmental contract mechanisms.”
But a look at the research on Twitter, such as the work done by the University of Iowa’s Philip Polgreen, Ph.D., who used Twitter content to predict flu outbreaks two weeks ahead of the CDC in 2009, reveals that the social media platform may be the most cost efficient and reliable means of prediction. The United Kingdom is relying on it now with norovirus — and it’s very likely that the United States will soon follow suit.