Queen Mary and PACE, just what is their case?

[Graham]

It's certainly worth a close read. There's lots of formal stuff, but they summarise many of the things that QMUL and Alem submitted, plus comments by "witnesses", then rubbish all the QMUL claims about the data being personal and it being possible to identify participants from the released information.

P 29: "The evidence before us is not clear but if QMUL are cherry-picking who analyses their data from within the recognised scientific research sphere to only sympathetic researchers, there could be legitimate concerns that they wish to suppress criticism and proper scrutiny of their trial."

 

[Chrisb]

You know that feeling you get when you have clicked "post reply" and have doubts as to whether it was entirely wise. Perhaps you don't, it may be just me. I'm sure you all have better judgment.

Do you think there is someone, somewhere, now thinking that perhaps the disparagement of our "world view", if there is such a thing, and if we have one, was a little unwise given the support it now appears to have?

Do you think he, or she, is thinking that the rather ostentatiously intimidatory tone was quite what should be expected in an objective legal document? "Only obeying orders" is not much of a defence.

One can only hope.

 

[TiredSam]

You know that feeling you get when you have clicked "post reply" and have doubts as to whether it was entirely wise.

Oh yes.

Do you think there is someone, somewhere, now thinking that perhaps the disparagement of our "world view", if there is such a thing, and if we have one, was a little unwise given the support it now appears to have?

I think the phrase "their world views" was speculatively thrown in there in the hope that it would appeal to the Information Commissioner's sense of snobbery and prejudice, on the off-chance that he was a member of the same old boy's network as the writer, and all it would take would be such a phrase to conjure up images of conspiracy theorists, protesters etc. It looks like they got lucky and scored a direct hit, so they didn't even have to explicitly refer to the fact that @Graham is a teacher, and we all know what left-wing trouble-makers they are.

This may be one of those posts where I end up wondering if it was entirely wise after clicking "post". Oh well.

 

[BurnA]

Just posting this here so new readers of this thread can follow up if they want:

New thread here:

http://forums.phoenixrising.me/inde...lease-anonymised-pace-data-16-aug-2016.46221/

 

[Snow Leopard]

They do not believe in it and therefore they attack it, often with obsessional attention to detail and a refusal to accept the integrity of the science.

The integrity of the science?

http://www.iflscience.com/chemistry/tips-spotting-bad-science/

1. Sensationalist headlines: Check
2. Misinterpreted results: Check
3. Conflicts of interest: Check
4. Correlation & Causation: Check (the mediation analysis studies)
5. Speculative language: Check
6. Sample size is too small: Sample size was fine.
7. Unrepresentative samples: Check. Oxford vs more rigorous criteria - the AHRQ sensitivity analysis in their GET/CBT meta-analysis reveals bias.
8. No control group used: Study was randomised, but strictly speaking, there was no true control group, in terms of controlling for all biases (updated due to comments below).
9. No blind testing used: Check
10. Cherry picked results: Substantial deviations from the published protocol and ignoring the lack of change of objective results is a red flag.
11. Unreplicable results: A change in subjective questionnaire answering behaviour is a reproducible finding. But this is hardly impressive when there are no change in objective functioning.
12. Journals & Citations: Unfortunately, Publishing in the Lancet does not guarantee high quality and should not lead to a free-pass.

Given the amount of red-flags above for "bad-science", why should we accept the integrity of the science?

 

[Woolie]

QMUL, you do need to be a bit more careful in what you say in these FOI responses. Your responses can be made available to anyone, and many don't show your institution in a good light.

 

[BruceInOz]

8. No control group used: A control group was used.

I don't think so. SMC differed too much in therapy contact sessions to be a control of the others and APT was supposedly an active arm. The 2011 paper does not claim to be a RCT but a parallel-group randomised trial. Another Check I think.

 

[Simon]

I don't think so. SMC differed too much in therapy contact sessions to be a control of the others and APT was supposedly an active arm. The 2011 paper does not claim to be a RCT but a parallel-group randomised trial. Another Check I think.

As I understand this, the "Standardised medical care (SMC)" group isn't a proper control group - such as a placebo pill group in a drug trial - because there is no 'placebo' therapy to control for effects such as therapist contact. All therapy groups also had SMC, eg for CBT the comparison was between the SMC-only group and the CBT-plus-SMC group.

(No, despite the authors' claims, APT isn't a pseudo-control group, despite the therapist contact etc, because of it's bizarre 70% rule that restricted patients' activity throughout the trial; the AfME pacing approach on which this is supposedly based currently has a 75% rule - but only to establish a robust baseline activity, before any increases.)