Nick Beeching

The same with the blood.

The antibodies take quite a long time to develop, sometimes up to four weeks or more after being infected.

So if you take the blood too early during the infection, you might not get a good result.

Yeah, I tend not to look at the boxes.

I'm a rather simple clinician.

So I think of it in terms of just sensitivity and specificity.

So a sensitive test means you don't miss anything.

You pick up everything.

You might pick up a bit too much.

And that's ideal where you're screening an infection.

You know, if you're testing people who might have an infection, you want a test that doesn't miss any, even if you then have to confirm it with a better quality test afterwards.

If you're looking to be absolutely sure about confirming that something has happened, then specificity is good.

In other words, you want it to be really accurate.

So if I had a blood test for antibodies...

and it was positive, I want that to be absolutely sure that that meant I'd had the infection and not had something else.

And antibody tests are a bit notorious for sometimes having cross reactions, especially early in the infection.

In practice, it's good to combine both.

But the most important thing for tests for the virus are essentially really good sensitivity.

It depends where you are and how many tests you have.

So in the ideal world, you would test lots of people at risk and you still test them for the virus and you might find it even if they have no symptoms at all.