Michael Sharpe skewered by @JohntheJack on Twitter

Jonathan Edwards said:
I am confused now. In these terms Godwin is right. An experiment that is unable to falsify a hypothesis is of not value as a test of that hypothesis.
Click to expand...
No, the theory and the hypothesis are two different things.

The theory - the theoretical framework which was used to derive the predictions with regard to CBT and GET - was that ME is caused by the mistaken belief that you are ill, which becomes further compounded by too much symptom focussing and also deconditioning due to fearful avoidance of activity.

The hypothesis is that patients in the CBT (or GET) trial arm will self-report less fatigue and greater physical function than a control group after 52 week.

The experiment tests the hypothesis. The hypothesis can potentially be disconfirmed. However, the experiment does not directly test the theory, because there is an added layer of assumptions between theory and hypothesis - which are about how you might best intervene to reverse the patient's fears and beliefs. If the assumptions you make here are wrong - for example, if you don't use the right technique, or don't give enough sessions or whatever - then you could get a null result, without your underlying theory necessarily being wrong.

So yes, technicially, a trial does not test a theory, not directly anyway. It tests a hypothesis derived from that theory. So all you can say about the behaviour/deconditioning theory from PACE is that the results didn't lend a lot of support for it. Even if all ther outcomes were negative, that would only falsify the hypothesis, it would not falsify the theory.

Edit: Sorry, @Jonathan Edwards, re-reading your post, I think this was exactly your point!
 
Last edited by a moderator:
Carolyn Wilshire said:
The theory - the theoretical framework which was used to derive the predictions with regard to CBT and GET - was that ME is caused by the mistaken belief that you are ill, which becomes further compounded by too much symptom focussing and also deconditioning due to fearful avoidance of activity.
Click to expand...

Is this falsifiable? Would you say that it has been falsified by any studies?

I know there was one study that found belief in viral cause vs belief in psychological cause were associated with similar outcomes.

Then there's the CPET literature which supports the belief that there exists an unusual physiological response to exertion.
 
There's also the fear avoidance theory, according to which it's more the belief that exertion causes deterioration, and that postexertional symptom exacerbation means harm and illness. The PEM and CPET literature seems to have explicitly disproved this, but that's just my view.

Whether the false illness beliefs and fear avoidance theories can be disproved, and how to do so, is an excellent question for PACE authors.
 
Carolyn Wilshire said:
The theory - the theoretical framework which was used to derive the predictions with regard to CBT and GET - was that ME is caused by the mistaken belief that you are ill, which becomes further compounded by too much symptom focussing and also deconditioning due to fearful avoidance of activity.

Th hypothesis is that patients in the CBT (or GET) trial arm will self-report less fatigue and greater physical function that a control group after 52 week.
Click to expand...

I would be cautious about characterising the underlying hypothetical model as a "theory" or "theoretical framework" in discourse.
 
Carolyn Wilshire said:
However, the experiment does not directly test the theory, because there is an added layer of assumptions between theory and hypothesis - which are about how you might best intervene to reverse the patient's fears and beliefs.
Click to expand...

Should they not first have gone back a stage further and looked for evidence which would refute the conjecture about fear of exercise being a contributory factor in the illness? There is plenty of evidence from the stories told on this site that, for many people, no such belief existed and that they continued exercising until having no alternative but to stop.

Whatever group of patients they were seeing there must have been some stories which should have caused doubts. But we did hear from Reviewer 2 that taking histories was not perhaps the strongest aspect of the team's practice.
 
chrisb said:
Should they not first have gone back a stage further and looked for evidence which would refute the conjecture about fear of exercise being a contributory factor in the illness? There is plenty of evidence from the stories told on this site that, for many people, no such belief existed and that they continued exercising until having no alternative but to stop.

Whatever group of patients they were seeing there must have been some stories which should have caused doubts.
Click to expand...

The reason they are not good at listening to patients is probably because they decided very early on that patients suffer from a functional somatic syndrome that involves false and/or delusional beliefs about the nature of their illness. What patients say then becomes very easy to ignore.
 
Sarah said:
I would be cautious about characterising the underlying hypothetical model as a "theory" or "theoretical framework" in discourse.
Click to expand...
Do you think the idea unworthy of the title theory? Do you prefer model, idea, just plain framework, or something else? I'm happy to go with whatever word you feel most suitable in this instance. The key point is how it differs from the study hypothesis (hypotheses).
 
Sarah said:
I would be cautious about characterising the underlying hypothetical model as a "theory" or "theoretical framework" in discourse.
Click to expand...

chrisb said:
Should they not first have gone back a stage further and looked for evidence which would refute the conjecture about fear of exercise being a contributory factor in the illness?
Click to expand...

Dredging my brain for anything I learned on my Science degree 50 years ago, I have a vague feeling that the word 'theory' was only applied to a situation once quite a lot of robust evidence had accumulated to support it as a theory.

If it was just an idea dreamed up by someone and not yet tested in any scientifically sound way, it was simply an idea, or a conjecture - something awaiting scrutiny, challenge and test.

My thoughts seem to be confirmed by these definitions:

Conjecture: an opinion or conclusion formed on the basis of incomplete information.

Scientific theory is a well-substantiated explanation of some aspect of the natural world, based on a body of facts that have been repeatedly confirmed through observation and experiment. Such fact-supported theories are not "guesses" but reliable accounts of the real world.
Click to expand...

On that basis, I think we need to stop giving the 'false illness belief' and 'fear avoidance' ideas of Wessely, White et al the status of a scientific theory. They are merely unfounded conjectures.

Edit: Crossposted with @Carolyn Wilshire.
 
Carolyn Wilshire said:
Do you think the idea unworthy of the title theory? Do you prefer model, idea, just plain framework, or something else? I'm happy to go with whatever word you feel most suitable in this instance. The key point is how it differs from the study hypothesis (hypotheses).
Click to expand...

Trish said:
My thoughts seem to be confirmed by these definitions:

Conjecture: an opinion or conclusion formed on the basis of incomplete information.

Scientific theory is a well-substantiated explanation of some aspect of the natural world, based on a body of facts that have been repeatedly confirmed through observation and experiment. Such fact-supported theories are not "guesses" but reliable accounts of the real world.

On that basis, I think we need to stop giving the 'false illness belief' and 'fear avoidance' ideas of Wessely, White et al the status of a scientific theory.
Click to expand...

Exactly this. It is insufficiently substantiated to be described as theory, and I wouldn't credit it as such.

Would "model" or "framework" prefaced by "hypothetical" be adequately distinct in your view? I think it's important to make it clear that the fear avoidance etc model of ME is itself hypothesis.
 
Sarah said: ↑
I would be cautious about characterising the underlying hypothetical model as a "theory" or "theoretical framework" in discourse.
Click to expand...

Carolyn Wilshire said:
Do you think the idea unworthy of the title theory? Do you prefer model, idea, just plain framework, or something else? I'm happy to go with whatever word you feel most suitable in this instance. The key point is how it differs from the study hypothesis (hypotheses).
Click to expand...

A narrative?
 
Sarah said:
Exactly this. It is insufficiently substantiated to be described as theory, and I wouldn't credit it as such.

Would "model" or "framework" prefaced by "hypothetical" be adequately distinct in your view? I think it's important to make it clear that the fear avoidance etc model of ME is itself hypothesis.
Click to expand...
I think we used "model" in the paper. Putting "hypothetical" in front would be confusing, for the very reasons we've been talking about there. It confuses the idea that inspired the hypotheses being tested with the actual fully operationalised hypotheses themselves.

In psychology, to avoid this kind of confusion, we tend to only use "hypotheses" for the fully operationalised version(s) of the prediction(s), the ones that are actually put to the test statistically.
 
Trish said:
On that basis, I think we need to stop giving the 'false illness belief' and 'fear avoidance' ideas of Wessely, White et al the status of a scientific theory. They are merely unfounded conjectures.
Click to expand...
Continuing my thoughts on this:
The PACE trial provided weak, subjective, transient evidence in support of the hypothesis that CBT aimed at changing false illness beliefs and fear avoidance would lead to recovery in CFS patients. Similarly it provided only weak, transient subjective evidence that GET is effective.

It therefore said nothing about the Wessely conjecture that false illness beliefs, fear avoidance and/or deconditioning are the main perpetuating factors of symptoms and disability in CFS. It simply said that the treatments designed to address these perpetuating factors were ineffective.

If the PACE trial had succeed in providing strong, objective and long lasting evidence of recovery with GET/CBT and not with APT/SMC, it would, I think, also have provided evidence in support of the underlying conjectures and moved it significantly towards becoming a theory.

In that sense, the PACE trial was, I think, designed not only to test the hypothesis that CBT/GET are effective, but also to provide support for the underlying BPS model and to move it more towards the status of theory.

Edit, sorry Carolyn, I keep crossposting with you. I'll shut up and eat my breakfast.
 
Trish said:
In that sense, the PACE trial was, I think, designed not only to test the hypothesis that CBT/GET are effective, but also to provide support for the underlying BPS model and to move it more towards the status of theory.
Click to expand...
Yes, if the hypotheses had been well supported, that would have provided positive support for the model. No question. It just doesn't work the same way for disconfirmation. Because people could fail to change for so many reasons - perhaps the therapists were not nice enough, there weren't enough sessions, their beliefs were too entrenched, etc.
 
You must log in or register to reply here.
Back
Top Bottom