David Tuller: Trial By Error: So What's Happening with the MAGENTA Trial?

So what's happening with the MAGENTA trial?

I’ll be in Bristol later this week for the CFS/ME Research Collaborative’s annual conference. I was not welcome last year, since I was at that point engaged in harshly criticizing the organization for its unwillingness to acknowledge that its deputy chair had falsely accused me of libel. This year, things have changed and both the chair and the new deputy chair have graciously welcomed me.

In any event, given the proximity of the conference to the University of Bristol, it seemed like a good time to take another look at some of the problematic research conducted on children at that august institution.

 

Bristol University have other problems with data
https://forbetterscience.com/2018/09/17/abder-kaidi-fraud-and-bullying-scandal-unravelling/

 

Yes. Seems to be several problems unravelling there..
There is a discussion on this thread, for those who wants to have a look
https://www.s4me.info/threads/epigr...-after-admitting-to-research-misconduct.5700/

 

@dave30th thank-you; you have made my day

 

With any luck, the journals will now be much more wary of publishing research like this. Especially when the flaws have been so nicely outlined by DT for the world to see!

 

Again .

 

Nice one. I think I made some mistakes with that MAGENTA registration that the blog avoided, and it got me thinking more about what we did know.

I still do not understand what's going on with MAGENTA though.

So there's still no protocol published for the full trial? No results for the feasibility study?

Some of the changes logged on the ISRCTN registry seem less than clear to me. I've tried to put in bold the dates of all the changes to make it easier to read:.

After those changes, the trial registration says:

Overall trial start date
15/01/2013

Recruitment start date
10/09/2015

I didn't understand that, so looked up the ISRCTN definition for 'overall trial start date':
"Overall trial start date: A study starts when you begin planning the design of the study and developing the protocol. The overall start date should precede the recruitment start date as the overall study period includes the recruitment period. Please give the anticipated or actual start date for the study in the format dd/mm/yyyy."
Ok - so, the overall trial start date would need to proceed prospective registration of the trial.

The ISRCTN definitions also say:
"Recruitment start date (WHO): The date, or planned date, of recruitment of the first participant to the study. If your study is registered before or on the recruitment start date, it will be listed as 'prospectively registered'. If your study is registered after the recruitment start date, it will be listed as 'retrospectively registered'."
So what's odd about this listing is that it started as the prospective registration of a feasibility study, and then on 27/03/2017 this was morphed into registration for the full study. Should this still count as prospective registration? One could argue that the registration was in place before recruitment began, allowing people to examine the changes to the registration logged on ISRCTN? That seems to defeat a lot of the advantages of prospective registration though as initially these were the only outcomes listed:

Primary outcome measures
Feasibility and acceptability of investigating GET in a randomised controlled trial measured after 1 year.

Secondary outcome measures
N/A

https://web.archive.org/web/20170228024303/http://www.isrctn.com:80/ISRCTN23962803

Also, what does anyone think this editorial note relates to? How are readers meant to know?:

"06/08/2018: Internal review."

 

I wonder if actometers were completely dropped, of it they're just downplaying them by not listing them as outcomes on the trial registration so that they can see what the results are before deciding what to do with them?

To me, it seems like dropping them would not be in Crawley's interest.

 

@Esther12

So this feasibility study game seems to be a blatant attempt to have the full trial (that the feasibility study gets rolled into) labelled as “prospectively registered”, instead of it carrying a more accurate descriptor of “retrospectively registered”.

Blatant gaming of the system surely? Why is this allowed? Or is it a loophole in the system not yet closed?

 

if the results from the feasibility study weren't good, why would she have wanted to keep them?

 

just in case anyone is reading this who has not been following previous debate here, thread
(as linked to by DT)
https://www.s4me.info/threads/magen...gers-and-pre-adolescents-esther-crawley.4808/

 

I think the 'loophole' will be the attitude of a chairman of a surveillance committee. Backed up by the attitude of one or more journal editors.

 

I don't think "loophole" is the right word. They're doing something they're not supposed to do. Prospective means prospective--as in, before the trial begins. What's been done here is not exploiting a loophole, it's violating scientific practice and getting away with it because no one is stopping you.

 

So are we saying they retrospectively wound the start back date, so they could retrospectively claim it to be ... prospective?!

 

I don't know enough about what the rules are for this sort of thing. It all seems very odd to me. It could be that other researchers think this is A-ok, or that they see it as appalling. I suspect their judgements will partly be founded on prior beliefs about Crawley/CFS!

If recruitment for the feasibility began on 10/09/2015 then a decision to drop actometer after the feasibility trial would have come when there was already so much attention on the 'objective vs subjective' outcomes issue with PACE, and PACE's worrying decision to drop actometers. In terms of her career, dropping actometers after they showed poor results for the feasibility part of the trial would surely be worse than just reporting poor results.

The PACE tribunal verdict was announced in August 2016, so probably before the feasibility trial was completed, so Crawley would be aware that even if she wanted to not release poor actometer results from the feasibility part of the trial, they could be forced out.

I feel like Crawley and PACE do lots of dodgy things that are close enough to being acceptable that researcher who aren't really paying attention will give them a pass. Dropping actometers after a null result just sounds like a needlessly big risk.

Admittedly, my arm-chair psychology is not a reliable source of information, and in the past I have been pretty shocked by some of the clearly unacceptable things they've done.

 

My impression is that the main players still thought they were untouchable until about November 2017. Up until that point Cawley was giving lectures on how to evade awkward questions about your research -presumably thinking she was succeeding.

 

I think the scenario is the feasibility started at time X, but it was then rolled into a full trial at a later date.

So, calling it a prospective trial at the time it was converted to full trial is wrong, because they are including the feasiblity study results.

Does that make sense?

 

I don't get that, because recruitment started in 2015 by all accounts, so at that point it was prospective. I'm not sure why they pre-dated it by two years.

 

and there was me just thinking it must be a typo.........