BBC Sounds, Music, Radio, Podcasts

Hello and welcome to More or Less, the show that sneezes, suspects, statistics into the

cook of an analytical elbow. I'm Tim Haafard. This week is all the soil in the world disappearing

or is it only all the soil in Sheffield. The very pleasant, hue-ferno-witting stall has

a pleasant, pleasant question for us and the answer to last week's Roman numerals puzzle.

First, as lockdown begins to lift, albeit in different ways and at different speeds in

different parts of the UK, the government is relying on testing and contact tracing programs

to prevent a second wave of Covid-19 infections. In England, the programme is called Test,

Track and Trace and it says that anyone with suspected coronavirus symptoms should isolate

for seven days, unless their test comes back negative in which case they can leave isolation.

So that approach relies on the tests being able to identify whether you're infected or

not, so how accurate are they? Jessica Watson is a GP and also a doctoral research fellow

at Bristol University, where she studies the use of diagnostic tests. So if I have Covid

symptoms and then I have a test to see if I've got the virus and the test comes back negative,

I think the natural assumption is, oh well I was mistaken and in fact I don't have it,

but it's not as straightforward as that is it? No exactly, it's never really quite that straightforward

with tests, so no tests are perfect. There's always the chance that the test might be falsely

reassuring or raising a false alarm as it were and with these coronavirus swabs, the particular

concern is the sort of false reassurance that sometimes occurs because we know that some people

who do have coronavirus will be missed by the swab tests. So do we know how often that happens?

If I've got a hundred people who actually have coronavirus and they all have this swab test,

do we know how many will falsely come back negative? It's a really good question, it's actually

...