And do you think that we're more sophisticated now as clinicians, as researchers, in interpreting meta-analyses than we were in the 1990s?

Yes, sophistication carries good and bad.

Sophistication means we do a better job in knowing exactly what we're about.

The other thing I don't like in many of the meta-analysis, they've become awfully statistically complex to actually read.

and the clinic the the the clinician who wants to be well informed they actually put a meta-analysis on one side as he tries to read his handcuffs couldn't they tell the story more straightforwardly and so one of my campaigns is to remember the reader you can do all the sophistication you like but then communicate in relatively straight wall forward statistical terms what you've actually found So too many meta-analyses have got into the methodological experts' hands who want to prove how clever they are in the wrong sense, and so you end up with somewhat indigestible articles.

Often the main result can be seen by a bit of stats on the back of an envelope.

Add up the numbers, it tends to give you the right answer a lot of the time across studies.

I wonder if you could tell us a bit about that.