PACE trial: fraud or incompetence?

[A.B.]

Err - how does that work.
It's clearly wrong - but how do you come to the conclusion that it proves intent.

The paper may provide evidence that White was aware that the calculation they used was misleading.

 

[alex3619]

PDW wrote a paper discussing SF36-PF and the use of standard deviation. He demonstrated he clearly knew it results in a very skewed result. According to David Tuller that was the point of the paper, or at least the point relevant here.

If someone knows that a particular method is mathematically invalid, and produces a very skewed result, then deliberately uses it or does not correct the problem later on, that has to be intentional.

I have not been able to get hold of the paper though, which is why I have repeatedly mentioned it. I don't even know its name.

This cannot be accidental if the lead investigator wrote a paper showing clear knowledge of the problem. It wasn't poor science.

 

[trishrhymes]

The only publication that might be relevant that I can find from Peter White in 2007 is this one:

(apologies if anyone has flagged it up already - it's doesn't sound like what you're looking for, but is interesting)

http://www.karger.com/Article/Abstract/99844

Is a Full Recovery Possible after Cognitive Behavioural Therapy for Chronic Fatigue Syndrome

Background: Cognitive behavioural therapy (CBT) for chronic fatigue syndrome (CFS) leads to a decrease in symptoms and disabilities. There is controversy about the nature of the change following treatment; some suggest that patients improve by learning to adapt to a chronic condition, others think that recovery is possible. The objective of this study was to find out whether recovery from CFS is possible after CBT.

Methods: The outcome of a cohort of 96 patients treated for CFS with CBT was studied. The definition of recovery was based on the absence of the criteria for CFS set up by the Center for Disease Control (CDC), but also took into account the perception of the patients’ fatigue and their own health. Data from healthy population norms were used in calculating conservative thresholds for recovery.

Results: After treatment, 69% of the patients no longer met the CDC criteria for CFS. The percentage of recovered patients depended on the criteria used for recovery. Using the most comprehensive definition of recovery, 23% of the patients fully recovered. Fewer patients with a co-morbid medical condition recovered.

Conclusion: Significant improvement following CBT is probable and a full recovery is possible. Sharing this information with patients can raise the expectations of the treatment, which may enhance outcomes without raising false hopes.
.....................................................................

I don't know the magic words to get access to the full paper, but I note that they claim 'full recovery' for 23% of patients, which is suspiciously close to the level they manipulated the data to achieve in PACE.

It would be interesting to know what criteria they used for both trial entry and 'full recovery'. Given that the PACE protocol is dated in the same year, they presumably used this trial to inform the criteria for PACE.

Can anyone wield the magic wand to get us the full paper?

 

[RogerBlack]

https://www.ncbi.nlm.nih.gov/pubmed/15789937 - I am unsure if this might be the paper in question, I have not read it in depth, as reading the history of PDWs submissions has drained me.

 

[Esther12]

Tuller has posted on this specific issue here: https://www.facebook.com/david.tuller.1/posts/10154194468391829?pnref=story

 

[Laelia]

The only publication that might be relevant that I can find from Peter White in 2007 is this one

Yes I think you might have found the correct paper here Trish, well done

Can anyone wield the magic wand to get us the full paper?

Anyone? (!)

 

[trishrhymes]

Yes I think you might have found the correct paper here Trish, well done
Anyone? (!)

Here's what David Tuller wrote on Facebook as @Esther12 linked to above:

I've split it up for ease of reading:
........................

''In my recent comment on the Dutch news site Observant, I noted that Dr. White, in a 2007 paper, acknowledged that using the formula for normally distributed populations yielded a distorted "normal range" threshold, but that the PACE team failed to include any such caveat in their 2011 and 2013 papers.

The 2007 paper, which Dr. White co-wrote with Dr. Knoop and Dr. Bleijenberg, among others, is called "Is a full recovery possible after cognitive behavioural therapy for chronic fatigue syndrome?"

Given that Dr. White obviously recognized the problems with the normal range calculations, it is unclear why he decided not to mention this problem in the PACE papers. Presumably, he didn't want readers to recognize that the "normal ranges" were anything but.

This kind of deceptive presentation of the data is typical of the PACE team, not to mention Bleijenberg and Knoop, who after all falsely claimed in a 2011 Lancet commentary that these bogus "normal range" thresholds provided a "strict criterion for recovery"--an astonishing, laughable and clearly bogus argument.

Here's the quote from the 2007 paper:

"In determining the threshold scores for recovery we assumed a normal distribution of scores. However, in the healthy population the SIP and SF-36 scores were not normally distributed.

Therefore one could argue that recovery according to the SIP8 has to be defined as scoring the same or lower than the 85th percentile of the healthy reference group.

In that case, the recovery rate using the definition of having no disabilities in all domains (i.e Recovery after Cognitive Behavioural Therapy for CFS scoring the same or lower than the 85th percentile on the SIP8) would decrease from 26 to 20%.

As we do not know the exact distribution of the SF-36 scores, we cannot control for the effects of violation of the assumption of normality." ''
......................

So Tuller was right - White knew his normal distribution assumption in PACE was nonsense.

 

[trishrhymes]

Hello, it's me again. I've just noticed this gem in the abstract I quoted in my previous post:

Conclusion: Significant improvement following CBT is probable and a full recovery is possible. Sharing this information with patients can raise the expectations of the treatment, which may enhance outcomes without raising false hopes.

So he's saying in 2007 that he expects to be able to influence patient outcomes by telling them there is a good recovery rate with CBT, which is exactly what he did in the PACE trial - knowingly influenced patients 'outcomes' for his favoured treatments. Deliberately.

 

[A.B.]

It's here

http://sci-hub.cc/10.1159/000099844

This is as simple as going to http://sci-hub.cc and pasting the URL, title, DOI into the search field.

Then maybe you have to fill in the displayed letters and hit enter to prove you're a person.

 

[trishrhymes]

Thanks, @A.B. I don't think I'll ever remember how to do that.

I've had a look through the paper briefly, and note particularly that in the 2007 study, they had far more criteria to be met to reach 'recovery' and the SF-36 physical scale required a score of 80 or more, which they say is one standard deviation below the healthy group mean of 93 (though of course, being a highly skewed distribution of non-linear data, this is meaningless). 59% of the group met that criterion after treatment.

I also notice that the measure that came out lowest in their recovery criteria was something called the SIP8 (sickness impact profile) where only 26% reached their recovery threshold. The combined recovery of patients meeting all criteria was 23%.

They did not use the Chalder Fatigue Scale, but used some other fatigue measures.

I suspect after this result they felt safe choosing SF-36 score of 85 as in their original protocol for recovery and 75 for improvement, expecting them to give nice high success rates. I can see why they didn't choose to use SIP8, as it served their nefarious purposes much less well!

I suspect it came as a nasty shock when they started to see the data coming in and found far fewer in any group were reaching SF-36 85 than they expected. No wonder they lowered it to the ridiculous 60.

One possible reason for this is that, as far as I can see, the 2007 study recorded their results immediately following the course of CBT, so the brainwashing effect had not had time to wear off. In PACE the outcomes were measured some months after the end of the treatment, I think.

 

[A.B.]

Thanks, @A.B. I don't think I'll ever remember how to do that.

Bookmark the post (small bookmark text at the bottom of the post). You can remember that you bookmarked it, right?

 

[trishrhymes]

Bookmark the post (small bookmark text at the bottom of the post). You can remember that you bookmarked it, right?

I did it!

Thank you kind sir.

 

[Esther12]

They cite that 2007 paper in the 2013 PACE recovery paper to make this inaccurate claim:

The other study used similar criteria
and domains for recovery (Knoop et al. 2007), but the
definition for normal range used was the more liberal
population mean-2S.D.rather than the more con-
servative 1S.D. that we used

So a cut off is 60 is more conservative than a cut off of 80?

Looking back at that recovery paper, I found their aims amusing:

The aims of this study were to: (a)define oper-
ationalized criteria for recovery on relevant domains,
(b) calculate the proportions of trial participants
meeting each of these individual criteria in each
treatment arm, (c) calculate the proportion of trial
participants meeting all the recovery criteria to pro-
vide a comprehensive and conservative definition
of recovery in each treatment arm, (d) compare the...

Anyone reckon they succeeded?

 

[alex3619]

I found this paper late last night, told someone about it in private conversation, then woke up about to paste it here ... and found the last bunch of posts. Doh. Good work!

I have yet to read the full paper.

 

[Gijs]

The same disciples of lord Wessely again, Knoop, Bleijenberg and White e.a. These masonic disciples feel the heat and trying to save their ass. A new CBT religion campaign is comming up bases on outdated data. It is fraud!

 

[Valentijn]

Not, alas the 'we can lock you up' commonly understood sense.

Well, not in a jail perhaps But it would suggest a bit of a disassociation from reality.

 

[slysaint]

Well, not in a jail perhaps

Broadmoor?

 

[Laelia]

I'll be honest? The fact that they MODIFIED criteria to fit the end result (not the desired test results), and all of the shenanigans to prevent data release, to attempt to slander critics, there's no way there were NO adverse affects (so that means they discredited the adverse effects OR HID THEM), etc, etc... I can't see how this can be tied to incompetence. I agree it could have started that way... but there is no way at this time there wasn't an intent to provide FALSE INFORMATION for the purpose of PROFIT.

Does that meet the criteria for Fraud?

[my bold]

It meets the criteria for scientific fraud/misconduct (not a legal issue but still has serious consequences).

 

[Laelia]

As I currently understand it any infraction on the scientific side is called scientific misconduct. Its the popular description by the public and press when they talk about fraud.

In scientific fraud, again my interpretation, there is intent to decieve.

[My bold]

Yes I think this is correct. Here is the used MRC definition of scientific misconduct (again). Sorry to keep repeating myself but there appears to have been a lot of confusion about this and I think it's important that we're all crystal clear on this.

The Medical Research Council (MRC) definition of misconduct and fraud (or a variation of the MRC code) is widely used. This code states the following definition:

The fabrication, falsification, plagiarism or deception in proposing, carrying out or reporting results of research or deliberate, dangerous or negligent deviations from accepted practices in carrying out research. It includes failure to follow established protocols if this failure results in unreasonable risk or harm to humans, other vertebrates or the environment and facilitating of misconduct in research by collusion in, or concealment of, such actions by others. It also includes intentional, unauthorised use, disclosure or removal of, or damage to, research-related property of another, including apparatus, materials, writings or devices used in or produced by the conduct of research. It does not include honest error or honest differences in the design, execution, interpretation or judgement in evaluating research methods or results or misconduct unrelated to the research process. Similarly it does not include poor research unless this encompasses the “intention to deceive” (MRC, 1997).[4] [My bold]

Fraud includes intent to deceive, and though I think there is a strong case for this it will be very hard to prove. That is not the case with scientific misconduct.

However, in an earlier post Alex, you wrote the above. I interpret what you have written here as saying that intent to deceive is not a necessary criterion for scientific misconduct. You might want to go back and edit this post to avoid any further confusion?

[Edit: We now know the above Medical Research Council (MRC) definition of scientific misconduct is innaccurate. Please see here for the correct MRC definition: http://forums.phoenixrising.me/index.php?threads/pace-trial-fraud-or-incompetence.50183/page-7#post-832589]

 

[alex3619]

However, in an earlier post Alex, you wrote the above.

This is an evolving discussion. The 2007 paper changes the basis for argument specifically for PACE. Intent to deceive is usually very hard to prove, but the 2007 paper may well be that proof for PACE. As I have been saying for years, you cannot demonstrate intentions without a paper trail or whistle-blower. This might well be the paper trail.

Scientific misconduct can be demonstrated without an intent to deceive, I think. Fraud cannot. In this quote "Its the popular description by the public and press when they talk about fraud" I was trying to point out that "fraud" is often how the more usual misconduct is discussed in the public and popular press. It may be that I should have explained that better.

I wonder if we can directly lodge a complaint with the MRC? If not, then who could?