Carl Heneghan

Therefore, let's move on.

And it's just list the things that make it moderate.

For instance, there was no blinding of the patient's

Without thinking around, how does this impact on the effect?

To what extent does it add to the uncertainty?

So I think that's a really important point, Duncan.

With the public, I tend to... And it's the same with everybody.

The first question of bias, I say, is this the right study to answer this question?

And if you look at lots of the news headlines, often you're looking at an animal study.

So immediately you're out.

The second question, then, is we discussed in the child kids.

Have you got all the available evidence to infer...

This is the true effect.

Have you got a systematic review or is this just an isolated study?

That's number two.

And then third, given you've got a clinically important outcome that will make a difference to you, within this study design, are there things that will undermine that result?

So they're the three steps, if you like, to trying to understand effects.

Yeah.

Competing interests declared.

Yeah, so that's really interesting.