How Twitter could save your life

Inane chat about runny noses, or pandemic predictor?

Back in 2010 AMC set up co-ordinated “zombie attacks” in major cities around the world to promote its zombie thriller series The Walking Dead. Gaggles of blood-dripping “walkers” invaded trains and lurched around landmarks like Big Ben and The Prado Museum. Just one small scratch, or, worse, a desperate, flesh-eating bite, and you would become a zombie too – in the drama, of course.

AMC’s most popular programme now pulls in over 12 million viewers per episode and has nearly 1.5 million Twitter followers, each obsessed with the dramatic, but scarily plausible, scenario of a true life version of blockbuster films like Outbreak, Contagion or 28 Days later.

But as Twitter continually proves itself to be such an adept viral tool, the sheer number of users – which is 500 million and counting – flocking to its pages could make it a hefty ally in the fight to contain such an outbreak. Twitter, it seems, may not only be the best place to send a  meme like the walking dead  ‘viral’, but also the perfect platform for stopping a virus dead in its tracks.

Twitter users react to current events and tweets contain real-time information about their perspective and location. If Lori Grimes, for example, had been on Twitter, could word have spread faster than The Walking Dead’s zombie outbreak? And could Contagion’s MEV-1 virus have been prevented if Beth Emhoff had tweeted about her supposed cold symptoms?

These questions might read like science fiction, but Professor Simon Hay at the UK’s University of Oxford believes there is a “revolution occurring” in the amount of public health data that is available through social media, particularly from Twitter.

While scientists have traditionally used mapping techniques to track outbreaks, it is just 4 per cent of infectious diseases that have been effectively mapped. New technology is required to improve results and Twitter could provide the answer.

In fact, Twitter has already provided geo-positioned information to inform scientists about public health. A study from the University of Iowa proved that content embedded in Twitter feeds relating to the H1N1 flu outbreak in 2009 allowed the tracking of “rapidly-evolving public sentiment” and “actual disease activity”.

By using Twitter's streaming application programmer's interface (API), the study explored public sentiment from 29 April to 1 June 2009 by identifying 951,697 tweets out of 334,840,972 that matched specified search terms, such as flu, swine, influenza, H1N1 and illness.

The second phase selected 4,199,166 tweets – which conformed to certain guidelines, such as they had to be in English and originate from the US – from eight million influenza-related tweets that included relevant keywords sent between 1 October and 31 December 2009. The study found that these Twitter feeds actually predicted outbreaks one to two weeks in advance of traditional surveillance.

Scientists are currently struggling to map the current outbreak of the H7N9 avian influenza virus in China – which is considered by the World Health Organisation to be a “serious threat” (126 have been infected to date and 24 have died), despite it not spreading through people as yet – so why isn’t Twitter’s data stream being utilised?

Could it be due to the lack of Twitter users in China? According to a programmer (@ooof) on the South China Morning Post blog, the number of live active Twitter users could be as little as 18,000. If this number was more, would scientists have been better able to predict this very real threat to our society’s health?

As an online flu detector exists in the UK, which has been created by a team at the University of Bristol through identifying keywords from Twitter’s geo-located content, then couldn’t similar programs be used to identify and predict other, more serious, infections?

Twitter has come a long way since it launched, when it attracted intense criticism from naysayers questioning why they would want to tweet inane information about an erupting spot or runny nose. But, in the battle against pandemic outbreaks, it is ironically these kinds of observations that could empower Twitter to become a sophisticated tool and actually be more than just a social lifesaver in the future.

Frances Cook is a freelance energy, transport and lifestyle reporter. She has worked for NRI Digital.

Getty Images.
Show Hide image

Could Labour lose the Oldham by-election?

Sources warn defeat is not unthinkable but the party's ground campaign believe they will hold on. 

As shadow cabinet members argue in public over Labour's position on Syria and John McDonnell defends his Mao moment, it has been easy to forget that the party next week faces its first election test since Jeremy Corbyn became leader. On paper, Oldham West and Royton should be a straightforward win. Michael Meacher, whose death last month triggered the by-election, held the seat with a majority of 14,738 just seven months ago. The party opted for an early pre-Christmas poll, giving second-placed Ukip less time to gain momentum, and selected the respected Oldham council leader Jim McMahon as its candidate. 

But in recent weeks Labour sources have become ever more anxious. Shadow cabinet members returning from campaigning report that Corbyn has gone down "very badly" with voters, with his original comments on shoot-to-kill particularly toxic. Most MPs expect the party's majority to lie within the 1,000-2,000 range. But one insider told me that the party's majority would likely fall into the hundreds ("I'd be thrilled with 2,000") and warned that defeat was far from unthinkable. The fear is that low turnout and defections to Ukip could allow the Farageists to sneak a win. MPs are further troubled by the likelihood that the contest will take place on the same day as the Syria vote (Thursday), which will badly divide Labour. 

The party's ground campaign, however, "aren't in panic mode", I'm told, with data showing them on course to hold the seat with a sharply reduced majority. As Tim noted in his recent report from the seat, unlike Heywood and Middleton, where Ukip finished just 617 votes behind Labour in a 2014 by-election, Oldham has a significant Asian population (accounting for 26.5 per cent of the total), which is largely hostile to Ukip and likely to remain loyal to Labour. 

Expectations are now so low that a win alone will be celebrated. But expect Corbyn's opponents to point out that working class Ukip voters were among the groups the Labour leader was supposed to attract. They are likely to credit McMahon with the victory and argue that the party held the seat in spite of Corbyn, rather than because of him. Ukip have sought to turn the contest into a referendum on the Labour leader's patriotism but McMahon replied: "My grandfather served in the army, my father and my partner’s fathers were in the Territorial Army. I raised money to restore my local cenotaph. On 18 December I will be going with pride to London to collect my OBE from the Queen and bring it back to Oldham as a local boy done good. If they want to pick a fight on patriotism, bring it on."  "If we had any other candidate we'd have been in enormous trouble," one shadow minister concluded. 

Of Corbyn, who cancelled a visit to the seat today, one source said: "I don't think Jeremy himself spends any time thinking about it, he doesn't think that electoral outcomes at this stage touch him somehow."  

George Eaton is political editor of the New Statesman.