Brain Retraining treatment for ME/CFS and Long COVID - discussion thread

[Jonathan Edwards]

NICE downgraded the evidence due to indirectness (not requiring PEM). It wasn't due to the quality of the evidence itself.

This is a non-sequitur statement.

NICE graded the evidence as poor according to GRADE. Part of that low grading was due to indirectness, but as Peter Barry made clear at the Round Table consultation before the Guideline was released (which I attended), the evidence was so poor anyway that the treatments would not have been considered cost-effective even without the indirectness factor.
So the downgrading relating to indirectness is not why NICE did not recommend CBT and GET.
The decision relied only on the quality and level of evidence itself.

 

[Jonathan Edwards]

But it didn't. It showed statistically significant improvements for CBT and GET.

Again non-sequitur. PACE failed to show that CBT and GET helped because the quality of the evidence was not up to doing that. Statistical significance is no use if there is a high risk of bias. Moreover the clinical significance of the difference was so tiny that the authors had to truncate the Y axis of the graph to make it look anything at all.

 

[Trish]

Even PACE failed to show that GET and CBT helped - so I think the "not shown to help" statement matches the evidence.

But it didn't. It showed statistically significant improvements for CBT and GET.

Yes, I'm aware. I think it's just important to be accurate in claims. The improvements were maintained at followup, and the control groups increased as well. There was no decrease in scores after the trial for CBT and GET.

It's also important when making claims about efficacy to put them in context. Just saying there was as statistically significant improvement on subjective questionnaires in the GET group compared to the SMC group at 12 months is not evidence of efficacy, for reasons we have already explained at length. The fact that this was maintained at follow up, as the authors like to claim is evidence of success is made nonsense by the lack of between group difference, as I'm sure you well know. Clinical trials of treatments should use the difference in effect between groups, not within a group.

I'm very familiar with ME/CFS, and I listen to patients. However, if you listen to patients you will see that psychological factors are important for many.

It would be helpful for us to understand your perspective in this discussion if you tell us whether you are writing from the perspect of a clinician listening to their patients, or a pwME reading patients stories on social media. I'm not asking you to tell us your name, that's up to you, but I'd like to know whether I'm trying to help a clinician understand better what may help or harm their patients, or a pwME to understand the basis of claims by their clinician to help them decide what treatment to try, a recovered pwME keen to share what they think helped them, or someone else who just enjoys provoking an argument.

 

[Utsikt]

It's frankly insulting and ridiculous to equate psychology with astrology and astrology and Scientology.

What is the foundation of psychology? It is speculations about how ‘the psyche’ or ‘mind’ or ‘self’ or whatever works. As @Jonathan Edwards has explained previously, we have absolutely no idea about any that. Zero.

So psychology essentially builds models based on unproven, and crucially, unverifiable assumptions.

So the very foundation of psychology is just shaky as astrology. Because it will always be impossible to falsify the axioms.

Actually there are multiple studies, as has been pointed out. I think what you mean is that you don't agree they are high enough quality.

It is not my opinion that they are not of high enough quality. It is a fact that they are not.

 

[Yann04]

cf. replication crisis for the results of the problematic methodological foundations of psychology.

Probably also been brought up, but there is a consistent history of psychology to psychologise diseases of unknown aetiology, until the aetiology is discovered. It’s a phenomenon that keeps on repeating itself. And when patients are told by doctors their problems are psychological, many believe them, whether or not it is true.

You should probably skim these posts https://mecfsskeptic.com/category/all-blog-posts-in-the-category-psychosomatic-medicine/

 

[Trish]

@dundrum and anyone else interested, Professor Brian Hughes book Psychology in Crisis is and enlightening read. It even has a chapter on the PACE trial.

 

[Yann04]

@dundrum and anyone else interested, Professor Brian Hughes book Psychology in Crisis is and enlightening read. It even has a chapter on the PACE trial.

Is there a detailed chapter by chapter summary of this somewhere? I’d very much like to read this but don’t have the energy to actually buy or read books.

 

[Utsikt]

Is there a detailed chapter by chapter summary of this somewhere? I’d very much like to read this but don’t have the energy to actually buy or read books.

I found a very very brief summary by someone here: https://www.researchgate.net/publication/331317113_Review_Psychology_In_Crisis

 

[Utsikt]

@dundrum and anyone else interested, Professor Brian Hughes book Psychology in Crisis is and enlightening read. It even has a chapter on the PACE trial.

Is there a detailed chapter by chapter summary of this somewhere? I’d very much like to read this but don’t have the energy to actually buy or read books.

Thread here:
https://www.s4me.info/threads/brian-hughes-2018-psychology-in-crisis.5535/

 

[Utsikt]

@Yann04 I have read the thread on the book now, and it seems like the review is slightly inaccurate. So apologies for that.

An important point is that the review claims that all of the issues with the PACE trial exaggerated the effect of GET. They did not. They created the illusion of an effect that did not exist in the first place. Nothing was exaggerated.

 

[dundrum]

It would be helpful for us to understand your perspective in this discussion if you tell us whether you are writing from the perspect of a clinician listening to their patients, or a pwME reading patients stories on social media. I'm not asking you to tell us your name, that's up to you, but I'd like to know whether I'm trying to help a clinician understand better what may help or harm their patients, or a pwME to understand the basis of claims by their clinician to help them decide what treatment to try, a recovered pwME keen to share what they think helped them, or someone else who just enjoys provoking an argument.

My goal is to help patients.

You're right, it is important to be accurate in claims. What you say is correct, but what people on here care about is whether the statistically significant difference between groups that existed at 52 weeks was maintained at long-term follow-up, and it was not. So what @Trish said was correct. The statistical significance that did exist did not persist.

What researchers have tended to do when the between-groups differences don't work out is focus on the within-groups differences, which are meaningless in terms of efficacy. They're only reassuring in terms of demonstrating that people did not report that GET/CBT made them worse on subjective outcome measures. Which unfortunately is not particularly reassuring, because people's self-reports are so vulnerable to being swayed, as Wechsler et al. 2011 shows. Wechsler et al. 2011 is mentioned above and is being discussed in this thread:

https://www.s4me.info/threads/activ...rvention-in-asthma-2011-wechsler-et-al.43655/

And what Wechsler et al. 2011 demonstrates so clearly is that the statistically significant differences in subjective outcomes that, if I've understood correctly, you see as evidence of some people improving from GET and CBT, tell us nothing about whether or not GET or CBT are effective for anyone with ME/CFS. But the lack of significant differences between groups in most objective measures do tell us that they're ineffective.

That's why Trish said "short term significance in subjective questionnaires".

Edit: Made a few edits for clarity.

What the PACE trial and the FITNET trial showed in long-term follow-up was that usual care eventually caught up with the treatment group, but much more slowly.

I agree that subjective outcomes are not ideal and introduce bias. The Wechsler trial is interesting, and does show that subjective outcomes are not useful in assessing asthma. However, are they useful in assessing a subjective outcome, such as depression from asthma? As for ME/CFS, we don't have any objective measures. What might be useful would be to measure activity for severe patients in a rehab trial, with a long-term follow-up. That is the only objective measure I can think of.

 

[Trish]

My goal is to help patients.

A laudable aim. So is mine. Can you tell us more about how you intend to help patients?

 

[Sasha]

I agree that subjective outcomes are not ideal and introduce bias.

In an open-label trial, why do you think that subjective outcomes are merely 'not ideal' rather than fatal?

You seem to be putting some sort of limit on the bias that they can introduce. Do you think that bias can turn a null result into a positive one?

As for ME/CFS, we don't have any objective measures. What might be useful would be to measure activity for severe patients in a rehab trial, with a long-term follow-up. That is the only objective measure I can think of.

Measuring activity for PwME was already thought of in the PACE trial, who used actimeters at the start of the trial and then went against their own study protocol and binned them. They would have provided an objective measure.

 

[Utsikt]

The Wechsler trial is interesting, and does show that subjective outcomes are not useful in assessing asthma.

That is not the only thing it shows. It proves that subjective outcomes can create an illusion of an effect with the same size as the effect of an actual treatment.

This means that we can’t categorically exclude the possibility that any other subjective outcomes can create a similar illusion, fully or partially.

Which in turn means that any subjective outcomes has to be considered to be inherently unreliable to at least some extent.

However, are they useful in assessing a subjective outcome, such as depression from asthma?

The fact that the thing you’re trying to measure is subjective by nature, doesn’t change the inherent unreliability of the subjective outcome measures. It has to be accounted for, even if it’s the best you’ve got. And there is no law that says that your best is good enough for science. So we have to entertain the possibility that it might not be good enough, period.

As for ME/CFS, we don't have any objective measures. What might be useful would be to measure activity for severe patients in a rehab trial, with a long-term follow-up.

I know we are speaking of hypotheticals here, but I want to emphasise that the first thing you should consider when designing an intervention trial is the risk of the intervention.

A rehab* trial for severe ME/CFS patients will never be ethical to conduct due to the risks involved. Unless, of course, the researchers are convinced that the intervention can’t possibly be harmful to severe patients under any circumstances. But that would make them entirely unsuited to conduct such a trial in the first place.
Edit: *based on the current prevalent rehab methodology.

 

[Utsikt]

What might be useful would be to measure activity for severe patients in a rehab trial, with a long-term follow-up. That is the only objective measure I can think of.

Off the top of my head:

• Step counters.
• Gyroscopic sensors to measure the orientation of the upper body and legs to try to assess orthostatic intolerance.
• Medication usage for symptom relief.
• Wages/Hours worked.
• Sleep measurements.
• Weekly time spent by carers.
• Tracking if specific ADL’s are performed or not.

None of these are perfect. But with rigorous methodology and neutral interpretations they might be able to tell us something?

 

[Evergreen]

What the PACE trial and the FITNET trial showed in long-term follow-up was that usual care eventually caught up with the treatment group, but much more slowly.

That's how the researchers interpret it. But because of the reliance on subjective measures in unblinded trials, and the dismal performance on more objective measures, the most justifiable interpretation is that people in the GET and CBT groups were sufficiently enthused that they temporarily changed how they completed questionnaires.

I agree that subjective outcomes are not ideal and introduce bias. The Wechsler trial is interesting, and does show that subjective outcomes are not useful in assessing asthma.

It's not people's asthma that made them respond as they did to the subjective outcome. It's their human psychology.

However, are they useful in assessing a subjective outcome, such as depression from asthma?

I'll let someone like @hibiscuswahine or @SNT Gatchaman answer regarding depression. But regarding whether subjective outcome measures are useful in assessing subjective symptoms like fatigue? The same human psychology would be at play, so no to exclusively subjective primary outcome measures, but yes to primary measures comprising both subjective measures and more objective measures.

As for ME/CFS, we don't have any objective measures. What might be useful would be to measure activity for severe patients in a rehab trial, with a long-term follow-up. That is the only objective measure I can think of.

Well, I wouldn't suggest severe patients go near a rehab trial until we have
(a) rehab being trialled that is significantly different from what has gone before,
(b) evidence from trials with multiple objective primary outcome measures that mild and moderate patients benefit (between-group differences, not within-group differences), and
(c) evidence that benefit lasts.
We do not have any of those. For those with mild and moderate ME/CFS, there are lots of other more objective measures like hours in work/education, or various tests physios use like how many sit-to-stands you can do in 30 seconds, but none as good as the forced expiratory volume test in asthma, because we don't understand the pathophysiology yet.

 

[dave30th]

What might be useful would be to measure activity for severe patients in a rehab trial

we had a rehab trial for severe patients, or at least homebound ones, called FINE. The PACE team kept touting it as their "sister trial"--but the PACE papers didn't mention it. It published null results for its subjective outcomes in 2010, a year before PACE--so PACE disappeared it, even though it had been billed as PACE's "sister trial." If that does not represent bad faith and constitute a form of research misconduct, I'm not sure what does.

 

[Jonathan Edwards]

And shed a little light on how health professionals really think:
"The bastards don't want to get better".

 

[Utsikt]

And shed a little light on how health professionals really think:
"The bastards don't want to get better".

Said by the bastards that don’t want to do better…

 

[dave30th]

Said by the bastards that don’t want to do better

The lead investigator of FINE was telling people that ME or CFS was like really bad jet lag. Since that's how those providing the care were essentially indoctrinated, I have some sympathy for some of their frustration at not seeing people improve. It doesn't justify hostility, obviously, but it's understandable to me. In other words, the nurses and others might have wanted to do better but were really being misled by the investigators.