Read on Twitter
The media is fairly consistently posting a graph that looks a bit like this to illustrate the current UK covid situation. This is a short thread about why I think it& #39;s important that they stop, and change their approach to illustrating the epidemic. [1/x]
![The media is fairly consistently posting a graph that looks a bit like this to illustrate the current UK covid situation. This is a short thread about why I think it& #39;s important that they stop, and change their approach to illustrating the epidemic. [1/x]](https://pbs.twimg.com/media/EicIeCqXYAIKlzd.png "The media is fairly consistently posting a graph that looks a bit like this to illustrate the current UK covid situation. This is a short thread about why I think it& #39;s important that they stop, and change their approach to illustrating the epidemic. [1/x]")
On surface reading, this shows the number of daily cases in the UK is not far away from where it was at the peak in March and April. I believe this is giving the covid-sceptical a false sense of security, while making the understandably anxious more panicked than needs be. [2/x]
To simplify this explanation, I& #39;ve made my own graphs, which roughly show the peak number of daily cases each month. Here& #39;s that same graph above, but adjusted to the new format. Yikes! We& #39;re almost back at peak infection rates! Right? [3/x]
![To simplify this explanation, I& #39;ve made my own graphs, which roughly show the peak number of daily cases each month. Here& #39;s that same graph above, but adjusted to the new format. Yikes! We& #39;re almost back at peak infection rates! Right? [3/x]](https://pbs.twimg.com/media/EicI8yGX0AAvOK8.jpg "To simplify this explanation, I& #39;ve made my own graphs, which roughly show the peak number of daily cases each month. Here& #39;s that same graph above, but adjusted to the new format. Yikes! We& #39;re almost back at peak infection rates! Right? [3/x]")
But there& #39;s something wrong with this graph. This shows "cases" - i.e. *confirmed infections* of coronavirus. But until May, we simply weren& #39;t testing enough for cases to be anything like representative of the actual number of cases. [4/x]
If you& #39;re covid-sceptical, this very much looks like a disease whose deadliness has drastically reduced, providing anti-restrictions lockdown. If you& #39;re anxious, it looks like almost as many people are being infected now as in April. Both of these are extremely false. [5/x]
We now know the infection fatality rate of covid to be roughly 0.4%. As testing has changed so much, we ought to disregard confirmed cases and use other data (such as surveillance studies and daily deaths offset by a month) to calculate the true number of infections. [6/x]
Do so, and you get a graph that looks more like this. Phwoar. Look at all those infections during the peak! Look how few there are now! That& #39;s why no one is dying! We& #39;re fine! Right? [7/x]
![Do so, and you get a graph that looks more like this. Phwoar. Look at all those infections during the peak! Look how few there are now! That& #39;s why no one is dying! We& #39;re fine! Right? [7/x]](https://pbs.twimg.com/media/EicJ8gHWkAIRVQ_.png "Do so, and you get a graph that looks more like this. Phwoar. Look at all those infections during the peak! Look how few there are now! That& #39;s why no one is dying! We& #39;re fine! Right? [7/x]")
Well, no, because one of the most important things to track is the trend. You& #39;ll have heard about something called the R number. That loosely tracks the speed at which an epidemic is growing or shrinking. Right now R=1.4 which means the epidemic doubles in size every week. [8/x]
So let& #39;s see what happens when we extrapolate that trend over the next couple of months and just check that we actually are fine, becau-- oh. [9/x]
![So let& #39;s see what happens when we extrapolate that trend over the next couple of months and just check that we actually are fine, becau-- oh. [9/x]](https://pbs.twimg.com/media/EicKcPfXYAE-XYc.jpg "So let& #39;s see what happens when we extrapolate that trend over the next couple of months and just check that we actually are fine, becau-- oh. [9/x]")
Put simply, if we keep going as we are, we will exceed the initial peak by the end of October. That means a thousand deaths a day again by the end of November. So. What does this all mean? [10/x]
I think it means it is time to be *concerned* but not to be *panicked*. There was probably as much covid in circulation at the start of February as there is now. [11/x]
We are stronger through knowledge, and time is on our side - but that won& #39;t be true for long. We have to take reasonably necessary precautions now, in order to avoid having to lock down society again in a month. [12/x]
Reasonable measures might include limiting gathering sizes, reducing hospitality hours etc. But we all have to play our part through social distancing, covering faces when needed, and reducing the number of contacts we have. If we can do that, we might be able to do this: [12/x]
![Reasonable measures might include limiting gathering sizes, reducing hospitality hours etc. But we all have to play our part through social distancing, covering faces when needed, and reducing the number of contacts we have. If we can do that, we might be able to do this: [12/x]](https://pbs.twimg.com/media/EicLYQjWsAMKOT5.jpg "Reasonable measures might include limiting gathering sizes, reducing hospitality hours etc. But we all have to play our part through social distancing, covering faces when needed, and reducing the number of contacts we have. If we can do that, we might be able to do this: [12/x]")
But if we leave it another month, we& #39;ll be very much back at the beginning of March again, and the only option will be a full lockdown, in order to make this happen: [13/x]
![But if we leave it another month, we& #39;ll be very much back at the beginning of March again, and the only option will be a full lockdown, in order to make this happen: [13/x]](https://pbs.twimg.com/media/EicLt6EWoAM0k5F.jpg "But if we leave it another month, we& #39;ll be very much back at the beginning of March again, and the only option will be a full lockdown, in order to make this happen: [13/x]")
In short, then, we are on the precipice and in the next month it will be crucial for the government to sort out the many problems with testing and contact tracing, decide and determine reasonable restrictions, and for us to change our behaviour to limit spread. [14/x]
If we can take reasonable steps to act now, it is genuinely possible for us to avoid either a huge spike in mortality, or a full March/April-style lockdown. But it does require everyone to play their part. [15/x]
In order for everyone to play their part, we need to be shown good, accurate data, in ways that we can understand, and which don& #39;t play into false narratives about the virus. And as the ones with everyone& #39;s attention, it is vital that the media takes a lead on that front. [16/16]
P.S. What graph would I be showing instead, if I were a newspaper or website editor? Probably something like this:
