How to spot a local coronavirus outbreak from data
FT data journalist John Burn-Murdoch looks at how the outbreak has unfolded in England and Wales and explains how data shows the difference between isolated clusters and community spread
Produced by Tom Hannen
Transcript
You can enable subtitles (captions) in the video player
This chart shows how the Covid-19 outbreak has spread through England and Wales since early March. Horizontally on the x-axis, we see the number of new cases per day per 100,000 people in every area, every local authority in England and Wales. Vertically, we're showing that as a rate of change, a rate of increase. So the higher up a point is, the more quickly it's new infections were increasing.
The lower down, the slower the rate of increase, or perhaps cases were actually declining. In early March, the coronavirus was spreading through England and Wales at a rapid rate. Numbers of cases that have been picked up are relatively low. But the rate of increase was extremely high.
Nationally, new cases were increasing week on week at a rate of seven or eight times. Moving later into March, that rate of increase began to taper off slightly. But as the increase fell, the numbers of cases grew. And soon, we had many communities in England and Wales with more than 10 new cases per day, per 100,000 people. Moving into April, that rate of acceleration continued to diminish slightly, but we were now moving towards the peak of confirmed cases in the community, with several areas now seeing more than 15 new cases per day, per 100,000 people.
Incredibly widespread infections. Moving from late April into May, both of those numbers started to go down as we moved past the peak. Numbers of new infections were decreasing, and therefore that rate of change was actually moving slightly negative. But you can see that we still had some communities hovering around on the right hand side there with elevated rates of infections per day.
Among those was Leicester, which became famous for its second lockdown in June. You can see here how Leicester's rates of infections was never especially low. It would maintain that background level of infection, which then increased through into early and mid-June. It has been joined then by more places that had very different outbreaks.
We see the meat processing factory outbreaks of the likes of Wrexham, Anglesey, and Merthyr Tydfil in Wales. But all of those were very short-lived. Note how the points jump out there to the upper right, but then quickly fly down into the lower left. Here we are now in July.
You can see that we now have that community spread going on in places still like Leicester, although it's down from its peak. And also places like Blackburn, Rochdale, and Bradford. The question now is whether cases will continue to fall in those areas, or whether that community spread might expand further.
One of the things we've seen here is how different places have very different types of outbreaks, different shapes of outbreak, we might say. We've seen the dots for places like Herefordshire, Anglesey, Merthyr Tydfil, and Wrexham, which shot into the upper right and then quickly back down into the bottom left.
This is what it looks like when you have an isolated cluster, something like an outbreak at a farm or a meat processing factory. Cases suddenly shoot up when a large number of people are suddenly detected in one fell swoop. And then as long as the outbreak hasn't spread into the community, cases fall very quickly as well.
These are better outbreaks to deal with, in the sense that they're small, they're isolated. There's only generally one place that has to be tested. If those outbreaks are contained, they come straight back down. That's in contrast to outbreaks like the one we saw in Leicester, and also the higher rates of infections we've been seeing in places like Blackburn. What's going on there is you have what we call community spread.
There's a slower increase, a more gentle slope upwards, as it were. A longer period with high infections and then a slower decrease. This is because the virus is spreading from person to person more widely in the community. There's no single, large outbreak.
These outbreaks can be harder to track and trace, meaning that they're more dangerous to the wider society. And there's a risk, if community spread is not controlled, that they can spread further and turn into regional or even national outbreaks, just like the one we saw across the whole of the UK, back in March and April.
The data used in this chart comes from Public Health England and Public Health Wales. The key thing to note here is that that data uses specimen date when we talk about timings. So that's the time, the day, when the swab was actually taken from somewhere. It's as close to real time information as we get on where new cases are being confirmed.
Every day, the FT is making a new piece of our best coronavirus coverage free to read. You can see all of our free coronavirus stories at FT.com/coronavirusfree, all one word.