It’s a question with a seemingly obvious answer – did two events where hundreds of thousands of people converged, sitting and standing in cramped terraces, really help spread the coronavirus?
This was the thrust of the question put to the government’s deputy chief scientific adviser, Angela McLean. Her response? A possible link between coronavirus cases in Liverpool and the local football club’s Champions League game against Atletico Madrid was an “interesting hypothesis”. The game, which took place on March 11, was attended by more than 52,000 people, including 3,000 from Madrid – a city which had shutdown schools just two days earlier in a country that had by then reported 35 deaths and 1,622 cases of coronavirus.
In that same week, another event took place that possibly shouldn’t have: the Cheltenham Festival. Over four days, 251,684 racegoers attended, with a crowd of 68,500 watching the Gold Cup on March 13. Amid media criticism over the event, the organisers would later cite Boris Johnson’s contemporaneous attendance at an international rugby match as a reason not to cancel the event.
When Liverpool’s match took place, the city had just six confirmed cases of coronavirus; now it has 1,210 and 246 deaths. The Gloucestershire Hospitals NHS trust, which covers Cheltenham, has 989 cases and has recorded 125 deaths, roughly double that in two nearby trusts at Bristol (58 each), and those covering Swindon (67) and Bath (46).
But while singling out these events is tempting, there isn’t yet the data to blame them for the spread of the virus.
It’s true that super-spreader events can occur – examples include the Zika virus potentially arriving in the build-up to the Fifa World Cup in 2014, or the cholera outbreak in Haiti following UN troops arrival there in 2005. “In the case of Covid-19, we do not have any evidence of this [having happened],” says Martin Hibberd, a professor at the London School of Hygiene and Tropical Medicine. “For many diseases, we typically do contact-tracing and genome sequencing for these situations, and if this had been done at these events, we might be able to infer something from that. In the absence of evidence, we can only speculate.”
Until this data emerges, these events should be considered on a par with a slew of other coronavirus-related missteps, like, arguably, the hundreds of flights that kept arriving from China throughout February. “In the end, there are likely to have been lots of communication visits between countries in Europe before [the UK entered] lockdown, and any of these could have been important,” says Hibberd.
Just looking at the figures is not of much use, either, says Kit Yates, a senior lecturer in the department of mathematical sciences at the University of Bath. For instance, if the per capita infection and per capita death rates in Liverpool and the surrounding regions show that the city has a higher per capita fatality rate (40 per 100,000 people) than some other conurbations (Greater Manchester, for example, at 32 deaths per 100,000 people), but lower than others (West Midlands at 52 per 100,000; Greater London at 44 per 100,000) and lower even than more rural areas (Cumbria at 45 per 100,000).
“The story is similar with case numbers,” he explains, “but even if Liverpool’s [figures] were significantly higher than other similarly populated regions’, it would be difficult to attribute this to one single even without more detailed information.”
A similar story holds for Cheltenham. “Again people have tried to argue that Gloucester’s case and death rates are higher than would be expected, but this isn’t borne out by the data at the moment,” says Yates.
To make such a claim you would need more detail: the time between that event and the disease taking off in the northwest, for instance, or reliable data from a poll of people who attended the event and later reported symptoms.
The most effective way of tracing the virus cases would come from analysing mutations in the virus’ RNA, a kind of signal in the genetic data of the virus.
“What we’re looking for is if any of the viruses from cases in the northwest and Cheltenham have been isolated and sequenced – because then we would expect to see different patterns in those viruses depending on where they come from,” says Andrew Preston from the Milner Centre for Evolution at the University of Bath.
Preston had been hoping to investigate the virus’s geographical spread with the help of the Gisaid international database of viral sequences, but he says that the database’s information is not granular enough. “We’ve looked at the publicly accessible data. It doesn’t include the location of the isolation of the virus, it’s simply at country level, which isn’t sufficient. The detailed location is there, but at the moment it’s not part of the publicly archived info,” he says. At least for the time being, there’s no way to chalk up outbreaks to specific failures.
Yet, just because we cannot say that a certain event made infections soar in a given location, this should not let the authorities off the hook.
“Both events were all the things you don’t want to do in terms of spreading disease,” says Preston “If you’ve got some people in that mix that are infected, you’ve got all the ingredients you would need to not only infect other people, but for them to take that infection away with them, and maybe introduce it into regions that otherwise didn’t have the infection at that point.”
Just the theoretical danger of the events – hundreds of thousands of people from all over the UK and from a country in partial lockdown mingling, coughing, and cheering – makes their incidence (at the very least) questionable. “Spain, of all places, and Madrid of all places in Spain – had significant infection already under partial lockdown at that point,” says Preston. “If [Madrid’s fans] were young, they may have had very mild symptoms or be asymptomatic and therefore maybe aren’t aware of their infection.”
The “open air” argument – that Covid-19 spreads less effectively during large outdoor gatherings – is scientifically sound, but in this case disingenuous.
“There’s a big difference between open air where people are running around in a park versus standing side by side jowl to jowl for five or six hours,” says Preston. “In Cheltenham, for example – it wasn’t all open air. They were inside bars. They were inside stadia. It was clearly very crowded.”
Will Bedingfield is a staff writer for WIRED. He tweets from @WillBedingfield
Coronavirus coverage from WIRED
😓 How did coronavirus start and what happens next?
❓ The UK’s job retention furlough scheme, explained
💲 Can Universal Basic Income help fight coronavirus?
🎲 Best video and board games for self-isolating couples