FOI request re PACE Cost-Effectiveness Analysis

[JohnTheJack]

In response to the Expression of Concern, the Statement of the Authors says:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0177037

During negotiations with the journal over these matters, we have sought further guidance from the PACE trial REC. They have advised that public release, even of anonymised data, is not appropriate. As a consequence, we are unable to publish the individual patient data requested by the journal.

I made an FOI request to the REC asking for a copy of that guidance and today received a response. The email they sent to the PACE people is attached. In the email to me they say:

Our response is as follows:
Please find attached the advice provided by the REC Chair. Please note from a transparency perspective the sharing of data would be the HRA’s normal recommended course of action. However as suggested in the attached email by the REC Chair, and from our discussions with other parties involved we understand the data may be made accessible via a secure site to other independent researchers wishing to analyse the data.

In light of this response, I have emailed Joerg Heber, asking that the anonymized, patient-level data be released to independent researchers, including Coyne, or the paper be retracted.


ETA: link to Expression and Statement

 

[Dolphin]

In light of this response, I have emailed Joerg Heber, asking that the anonymized, patient-level data be released to independent researchers, including Coyne, or the paper be retracted.

I understand why someone might suggest this. However, this paper is useful in showing how ineffective the therapies were.

 

[JohnTheJack]

I understand why someone might suggest this. However, this paper is useful in showing how ineffective the therapies were.

I'm hoping that enough pressure can be put on them to share the data.

 

[ME/CFS Skeptic]

Many thanks to John.

I have a technical problem however: can't seem to open the attached document. My adobe is nonetheless up to date. Am I the only one with this problem?

 

[Trish]

I couldn't open it either. It told me I had to download Adobe.

 

[JohnTheJack]

The file is a bit odd. I have done a screenshot of the body of the email sent in reply to the request for guidance. The original request has been redacted.

 

[oldtimer]

Here it is in full.

 

[JohnTheJack]

Here it is in full.

Thanks.

 

[Trish]

I wonder if they were aware that a lot of the PACE data has been put in the public domain and no harm to anyone has ensued. I don't understand why they think this particular data should be treated any differently.

 

[Barry]

I have a technical problem however: can't seem to open the attached document. My adobe is nonetheless up to date. Am I the only one with this problem?

I couldn't open it either. It told me I had to download Adobe.

If you save it first then it can be opened from disc. I had the same problem.

 

[Esther12]

I understand why someone might suggest this. However, this paper is useful in showing how ineffective the therapies were.

That data would still be available so it's hard to believe that withdrawal of the paper would be unhelpful. Citing their poor results and then pointing out that the paper was withdrawn due to their refusal to abide by the data sharing requirements they'd signed up to sounds okay to me.

 

[Dolphin]

That data would still be available so it's hard to believe that withdrawal of the paper would be unhelpful. Citing their poor results and then pointing out that the paper was withdrawn due to their refusal to abide by the data sharing requirements they'd signed up to sounds okay to me.

I doubt reviews of the field would include retracted studies. The point of retracting studies is I presume to remove the findings from being used as a basis for anything.

So patients might refer to them in lay literature, but I think the standing of the findings would be lowered greatly and I think people would be reluctant to allow them be referred to in peer-reviewed publications.

 

[Esther12]

I doubt reviews of the field would include retracted studies. The point of retracting studies is I presume to remove the findings from being used as a basis for anything.

That's a fair point, but the null results for employment kept being ignored anyway. If the paper was retracted due to the researchers refusal to follow data sharing requirements, I suspect that a thorough review would still be able to note the results and the reasons for withdrawal.

I was going to say that it would be odd to insist people just pretend these results from the largest CBT/GET trial did not exist, but then I remembered that this seems to be what has happened with the 2.5 year LTFU employment data that has still not been released.

IMO, regardless of any downside, retraction would be useful as a political signal that the PACE researchers could no longer get away with having their powerful friends bending the rules for them, and that we were moving towards a period of greater transparency and accountability.

 

[Trish]

I understand why someone might suggest this. However, this paper is useful in showing how ineffective the therapies were.

I think we need to stick to our principles. If a paper is based on bad science, which all the PACE papers are, and if the authors of a paper refuse to comply with the publisher's policy of open data, then it should be retracted, no matter how useful to our case some of its results may be.

 

[ladycatlover]

If the paper was retracted due to the researchers refusal to follow data sharing requirements, I suspect that a thorough review would still be able to note the results and the reasons for withdrawal.

The difficulty is that one wonders how thorough any reviews are.

 

[Barry]

or the paper be retracted.

I understand why someone might suggest this. However, this paper is useful in showing how ineffective the therapies were.

I'm hoping that enough pressure can be put on them to share the data.

On the one hand I agree with @Dolphin that retraction might lead to a useful paper being "lost" to further studies, but I also see that there is a point of principle here, and we cannot really be asking for retraction of other ME/CFS research due to ethical oversights, yet skipping ethical oversight where it suits us.

I think there is also the notion of precedent here. Given who the authors are, the likelihood of retraction is nigh on nil I would think, but if it were retracted then that would be a real precedent being set - a Sharpe paper being retracted! To that end I think the loss of publication of the paper would be hugely beneficially outweighed by the loss of Sharpe to the published paper, if you get my drift. I cannot imagine MS wanting to be pushed into that corner, so I fully support the notion of doing so, especially if it ended up with data being released.

 

[Dolphin]

I think we need to stick to our principles. If a paper is based on bad science, which all the PACE papers are, and if the authors of a paper refuse to comply with the publisher's policy of open data, then it should be retracted, no matter how useful to our case some of its results may be.

Much of the bad science in the PACE Trial is common to non-blinded studies which aren’t going to disappear any time soon. Other PACE Trial papers have worse flaws e.g. normal range calculations in Lancet paper; ridiculous revised recovery criteria in recovery paper.

 

[Trish]

Much of the bad science in the PACE Trial is common to non-blinded studies which aren’t going to disappear any time soon. Other PACE Trial papers have worse flaws e.g. normal range calculations in Lancet paper; ridiculous revised recovery criteria in recovery paper.

Good points. I'm hoping the request for retraction will push the PACE authors to release the data to independent researchers.

 

[Sly Saint]

Don't know if this has been posted already or not.
This is from the MRC in Aug 2018:

"The MRC has recently started (with other research funders) a pilot of a new facility for academic researchers to share data via the Clinical Study Data Request (CSDR) web portal. We hope to include the PACE trial on the CSDR portal within the next 6-12 months. This will allow researchers to apply to the CSDR Independent Review Panel to gain access to the dataset for their own analyses."

https://mrc.ukri.org/news/browse/criticism-of-the-pace-trial/

 

[ladycatlover]

"We hope to include the PACE trial on the CSDR portal within the next 6-12 months. This will allow researchers to apply to the CSDR Independent Review Panel to gain access to the dataset for their own analyses."

And how independent will this Independent Review Panel be? Will they allow people like Tuller and Geraghty to get access? I really don't see them letting @dave30th have access - would be too embarrassing for certain people. Of course I hope I'm proved wrong. It will certainly be interesting to see what transpires!