Tracking Covid numbers at this stage of the pandemic

It’s being called “Super Saturday” here in the UK. Pubs, hotels, hairdressers, and other hospitality industries are being allowed to re-open. The Foreign Office has dropped its advice against non-essential travel to a long list of countries who seem to have dealt with the virus as well as (or better than) we have. It’s not fully back to normal yet, but it’s quite a bit closer today than it was yesterday.

This is possible because we’ve come a long way in suppressing the virus since the peak of the pandemic. 20 April was the most deadly day in the UK, with 1,173 deaths recorded. Compare that day to more recent figures, with 1,050 deaths recorded in the last nine days combined (24 June to 2 July). And before the government’s large downward revision in number of cases mucked up my spreadsheet earlier this week (bye bye trendlines!), the seven-day moving average of cases has broadly been going down, with fewer than 1,000 confirmed cases a day on average over the previous seven days (the moving average on this measure reached a peak of nearly 5,500 cases a day in mid-April, but given the poor state of testing infrastructure at the time the actual number was probably several times higher).

Image for post
Image for post
Daily deaths with confirmed COVID-19 in the UK, since the first such death was reported

This week saw the start of the first local lockdown related to the pandemic, with the city of Leicester, and some of the villages on its immediate outskirts, forced to close its non-essential shops and suspend its planned hospitality re-openings. In announcing the lockdown, Health Secretary Matt Hancock told the House of Commons that Leicester had 10% of all confirmed UK cases in the week to 30 June, which would suggest nearly 100 new cases a day within the city.

At the moment, we are seeing the seven day average of cases decline by more than 100 each week (well, we were before the revision mucked up my spreadsheet a few days ago), and so the nationwide totals average can, at the moment, continue to go down despite the local outbreak in Leicester. However, the trendline generally points to the number of daily cases halving from one month to the next. This means that, in the next month or so, we may reach the point where another local outbreak, of a similar absolute size to what’s happening in Leicester, will become big enough to affect the national trend.

I’ve consistently said that, when looking for trends, once should look at seven-day rolling averages to smooth out day-to-day fluctuations. But when even these rolling averages start to go up (which they will before the summer’s out), more detailed analysis will be essential. If we find that it’s driven by a small number of local areas, then a nation-wide rollback of lockdown easing would not be justified. But if there is no obvious local area driving the increase, then such a rollback would become necessary.

So as we look to enter this new stage, as the case numbers (and thus the denominators in many calculations) becomes lower and lower, we need to be more and more careful with our analysis of data, and ensure that we can work out what is a local spike, and what truly is a national trend.

Image for post
Image for post

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store