Bristol - New Network - Prof Munafo

[Suffolkres]

Cochrane not Cockcroft?

I thought the recent Cochrane spat meant we had a new collaborative.....!
https://blogs.bmj.com/bmjebmspotlight/2018/09/16/cochrane-a-sinking-ship/

Sounding a tad familiar....?

Cochrane’s sinking ship

The events that have unfolded in the last few days have consequences for Cochrane far beyond dealing with the public embarrassment of losing more than a third of its Governing Board.

Much of Gøtzsche’s scientific work at the Nordic Cochrane Centre, has focused on exposing the flaws in clinical trials and the undue influence of the drug industry on medical research.

In addition, there are the issues raised in a recent editorial, co-authored by Dr Tom Jefferson from Centre for Evidence-Based Medicine, Oxford. It explains the problems behind the reliance of data from published journal articles, many of which are likely to contain ‘unfathomable bias’.

“We know that the biomedical journals publish articles which are neutral at best, but are mostly positive and tend to emphasize benefits and downplay or even ignore harms,” says Jefferson.

 

[Graham]

Cochrane not Cockcroft?

Oh dear, oh dear, oh dear! Cockroft carried out an important review in education!

 

[MEMarge]

I would probably put something like "flawed trial" rather than just "bad".

Feel free to ignore. I have minimal experience in these things and we would probably all write something slightly different.

Thanks again for following this up.

 

[TiredSam]

In general, I prefer to focus on broad issues rather than specific studies. This is partly to ensure that I don’t get mired in any particular debate ...

Sounds like the Wessely approach. Details and committing yourself can be so tiresome.

 

[Graham]

I've added the amendments, and sent it. An automatic reply said that he would be out of his office until 1st October.

 

[MSEsperanza]

Sadly, with very few exceptions, the only UK researchers who feel qualified to comment are the authors themselves.

I've added the amendments, and sent it.

 

[Graham]

Despite being out of office, Prof Munafò has replied:

Dear Graham,

Thanks for your feedback. I’ll think about what strategies might be effective, and take on board what you say about the Network identifying poor quality work. As we grow that might be something we have the capacity to take on. I would also say that Cochrane reviews *do* attempt to weight by quality, albeit crudely. The problem is that different people have different views of what constitutes quality! Good luck with your efforts.

All the best,

Marcus Munafò
Professor of Biological Psychology

I have sent back a short note, thanking him.

 

[Skycloud]

Thank you for this Graham!

I hope Prof. Munafò is genuine in his response.

 

[Sasha]

Nice job, Graham - good that he's thinking.

 

[Barry]

Despite being out of office, Prof Munafò has replied:

Dear Graham,

Thanks for your feedback. I’ll think about what strategies might be effective, and take on board what you say about the Network identifying poor quality work. As we grow that might be something we have the capacity to take on. I would also say that Cochrane reviews *do* attempt to weight by quality, albeit crudely. The problem is that different people have different views of what constitutes quality! Good luck with your efforts.

All the best,

Marcus Munafò
Professor of Biological Psychology

I have sent back a short note, thanking him.

Excellent. Unless the reply is insincere (hopefully not) then I believe you may have helped Professor Munafò gain a crucial insight - that he should not blindly accept all that he is told by those around him. If so then that is a very significant step forward.

 

[Jonathan Edwards]

I think Munafo has listened. And not shut the door. My impression is that he might have said more but that he also has to gauge what he says in public. A useful exchange.

 

[MSEsperanza]

bump

 

[MSEsperanza]

A propos, Prof Munafo is on Twitter:
https://twitter.com/MarcusMunafo

 

[Suffolkres]

A propos, Prof Munafo is on Twitter:
https://twitter.com/MarcusMunafo

Interesting tweet...perhaps we should send it to Bath.......

Marcus Munafò Retweeted

Guillaume Rousselet
@robustgar
·
Aug 29A reminder that the Declaration of Helsinki makes pre-registration mandatory: 35. Every research study involving human subjects must be registered in a publicly accessible database before recruitment of the first subject. https://wma.net/policies-post/

 

[rvallee]

Interesting tweet...perhaps we should send it to Bath.......

Marcus Munafò Retweeted

Guillaume Rousselet
@robustgar
·
Aug 29A reminder that the Declaration of Helsinki makes pre-registration mandatory: 35. Every research study involving human subjects must be registered in a publicly accessible database before recruitment of the first subject. https://wma.net/policies-post/

I do love the fact that Sharpe once stated that the reason The Lancet fast-tracked their paper is because they pre-registered, which is a norm as has been reminded here, even though they strongly deviated from the registered protocol. Also: "equipoise", sure, whatever. The principal author and another author were directors in a company with the intent of promoting the very same model and about half the authors had written books and promoted the treatments as safe and effective for years. Completely absurd.

Schrodinger's trial: pre-registered but also pre-registration is bothersome when you prefer a certain outcome that aligns with your expectations, as otherwise "no one would have recovered" *cue audience cackling while millions are suffering and many are dying needlessly from zingers like this*.

 

[Esther12]

I do love the fact that Sharpe once stated that the reason The Lancet fast-tracked their paper is because they pre-registered

I don't remember that. Can you remember where that was?

 

[rvallee]

I don't remember that. Can you remember where that was?

It was in a tweet but I can't search because he blocked me.

Someone isn't blocked should be abke find it with the terms "lancet" and "fast-track", or something like that.

 

[Esther12]

It was in a tweet but I can't search because he blocked me.

Someone isn't blocked should be abke find it with the terms "lancet" and "fast-track", or something like that.

Thanks - here you go:



Also relevant:

 

[Adrian]

Thanks - here you go:

The comment does say regisitered the protocol with the Lancet rather than preregisted. But of course the 2011 paper ignored the published protocol as this was rewritten with their stats analysis plan.

 

[rvallee]

Thanks - here you go:



Also relevant:

That's the one!

Would be interesting to know what % of pre-registered papers are fast-tracked. I'm gonna go ahead and bet on no relation whatsoever between pre-registration and fast-tracking.

Although technically, PACE wasn't pre-registered with the whole "not following what was registered". Not following a pre-registered plan is the same as not registering the actual plan. That's like making a prediction after the fact and claiming it was an accurate prediction because an inaccurate prediction was made before the event. Complete BS.