Harlan Krumholz

You've got this cytokine storm.

You want to reduce the inflammation.

Steroids do that.

But actually, it seems like the evidence is showing you it's a bad idea.

Whereas in the antibiotic issue, I'm not aware of any evidence that shows you it's beneficial or harmful in a viral infection.

We worry that if it's

And what happens there is we're worried you're giving antibiotics when you don't need them and you get the side effects.

But what's happening here is people are thinking experience, influenza, secondary bacterial infections, hospital-acquired pneumonia, where we're going to treat until we certainly haven't got pneumonia.

So what you might see from a clinical experience is people giving 48 hours of antibiotic treatment in the middle of this high temperature, awaiting blood cultures and other information that can allow you to then rule out it's a bacterial infection.

Just to say, though, it's really interesting.

You think, well, there's going to be no evidence here.

And in one of my advisory roles, I'm an advisor to the World Health Organization's International Clinical Trials Registry platform.

And Ghassan Karam, who runs that, sent out a tweet yesterday and he sent me an email saying, look, we have a novel coronavirus registry.

They put all the trials that are ongoing together.

There are currently 81 trials on the platform in recruitment.

So that's quite surprising.

And the speed at which people are operating is quite something.

So one of the things we can do is there's a whole base of array of treatments that I'd have to go across, but...

Some people are still thinking about low-dose steroids.

Some people are using the antiviral treatments, remdesivir, which is a novel antiviral treatment, aerosol inhalation on there, all sorts of studies of agents, but also some of the early one in prediction systems.