“We’re now doing about 240,000 antigen tests a day, at the moment… 240,000 is the latest figures… This is the capacity in the system.”
Nadhim Zahawi MP, 17 September 2020
“The government’s own figures, the latest figures, are that 81,000 people are being tested every day. That’s the government’s latest figures. Not 240,000. 81,000.”
Fiona Bruce, 17 September 2020
There seems to be widespread confusion about the difference between the number of people being tested for Covid-19 and the number tests actually available, or being performed.
In particular, many prominent people in the media and in politics are wrongly comparing the number of people newly tested in England (reported in the weekly Test and Trace statistics) with either the testing capacity for the whole UK, or the number of tests actually performed (both from the Coronavirus Dashboard).
Crucially, that figure for people “newly tested” won’t include most people who’ve been tested more than once since the Test and Trace system began operating.
This mistake appeared in the Sunday Times at the weekend, Angela Rayner repeated it in Prime Minister’s Questions on Wednesday, and Fiona Bruce made the same mistake during Question Time on Thursday. Ms Bruce’s mistake was later repeated by Femi Oluwole, and other prominent Twitter users.
In Parliament on Thursday, Jacob Rees-Mogg MP also confused tests performed with people tested when he said that, “nearly a quarter of a million people a day can be tested”.
We don’t know the true figure for how many people are being tested for Covid-19 in the UK each day, because the government does not publish it. We asked the Department for Health and Social Care if they could provide us with this figure, but they did not.
What is everyone getting wrong?
There is currently more demand for Covid-19 tests in the UK than the system can meet.
The official Coronavirus Dashboard publishes a “lab capacity” figure, which reports how many tests the labs say they can process each day. It also publishes a “tests processed” figure, which shows how many tests they did process.
Even at this time of excess demand, the labs have only once processed as many Pillar 1 and Pillar 2 tests in a day as they expected to be able to. (Pillars 1 and 2 of the testing system are the ones that cover swab tests to see if someone is currently infected with the coronavirus, and account for the bulk of tests carried out.)
As we have said before, this suggests that the capacity figure may not really tell us how many tests can happen in practice. The government itself also admits that “operational issues at labs can result in labs performing below capacity”.
The most tests that have been processed in a single day was 238,640 on 12 September, so Mr Zahawi’s figure was broadly correct (although the latest figure for tests processed is a bit lower, at 223,146). The latest figure for claimed capacity is currently just over 240,000.
However, this does not necessarily mean that we should expect to test 240,000 people, because some people are tested more than once for a variety of reasons. (For example, Public Health England currently recommends that labs test borderline positive cases a second time). The fact that the number of tests are different from the number of people tested is not a new issue; we first pointed it out in April.
What we don’t know is how many tests are being carried out on average for every person tested, because the government doesn’t publish figures that would let us work that out.
Currently, there is no national figure for the total number of people being tested each day either. The closest thing we have is the number of people “newly tested” each week in England, which is published in the Test and Trace statistics. This was 571,400 in the latest figures for the week up to 9 September, equivalent to about 82,000 a day.
Because this figure only covers England, it misses the people who were tested in the Scottish, Welsh or Northern Irish system. And because it only counts people who were newly tested in that week, it also probably misses most people who have been tested before, which might for example include frontline health workers, or staff and residents in the care system. (There is an exception for people who test positive for the first time, after testing negative before. These people are moved to the week of their first positive test.)
Without better data, we don’t know how much difference it would make if we included all these people. The UK’s testing system could be testing only slightly more than 82,000 people each day—or it could be testing many more.
Right now, we can only say that we can’t compare the “tests processed” or the "testing capacity” with the “people newly tested” figures, because they measure different things.