IiME letter to Mark Baker (NICE) re: CBT & GET as recommended treatments

In their letter, Peter White et al state: “The PACE trial paper refers to chronic fatigue syndrome (CFS) which is operationally defined; it does not purport to be studying CFS/ME”.


The sentence continues by stating that the PACE Trial studied: “CFS defined simply as a principal complaint of fatigue that is disabling, having lasted six months, with no alternative medical explanation (Oxford criteria)”.

http://www.meactionuk.org.uk/Hoopers-initial-response-to-PDW-letter.htm

 

That confuses me.

 

I'm not even sure that they didn't use an unpublished criteria called the London criteria in the PACE trial.

We really need some input from PACE trial geeks here (and not just the data mining types) because the Oxford criteria issue is much more deceptive than some people may think.

 

You know me. My mind wanders. All this talk of Oxford criteria made me inquisitive about the two others, along with Peter White, thanked for financial support for the conference establishing the criteria.

They were:

1. Duphar Pharmaceuticals. They appear to have changed their name to Solvay Healthcare. The first link I found was to a "Pharmaletter" announcing the name change in 1995. The sidebar has a prominent link to an article "Mood disorders market set to grow to $9.5 Bill by 2004."

2. Professor Michael Gelder of the Department of Psychiatry, Oxford University whose claimed specialties are "psychiatry, cognitive behavioural treatment, clinical trials". How many of you were aware of his interest in ME. Or CFS, for that matter.

 

It's much easier to understand that PACE is bad if you know how easy it is to get improved responses to health questions.

PACE is not the only trial like this, and NICE will find it difficult to accept that the benefit of many other treatments is in doubt as well because of similarly bad methodology. Yet that is what they must accept. Perhaps talking about a systemic problem and how the time has come to demand higher standards in psychotherapy will work.

 

Yes but if you are studying the most popular citrus fruits and you use bananas as your criteria because you work for Fyffes that's flawed.

 

Oxford criteria on their own may not automatically label PACE bad science, however, NICE is not solely about good or bad science, it is about the application and interpretation of science.

If they want to apply PACE findings to ME
then surely this must be considered as being poor interpretation or irrelevant even.

People with joint pain do not automatically have RA, the same way as people who are fatigued for more than 6 months don't automatically have ME.

Just because a group of psychs from Oxford come up with their own definition doesn't mean their work applies to any other definition even if it is possibly wider.

For example if PEM is a mandatory symptom for ME, then Oxford studies do not comply. It doesn't matter how wide they are.

If NICE want to have a section for unexplained fatigue lasting 6 months or longer then they could consider PACE results, but obviously even then the results are negative.

I agree about not fighting any issue which isn't rock solid, but I am wary of compromising before it begins.
There may be a lot of compromising to be done later, we shouldn't surrender some of our arguments before the off.

It's not like this is contentious, the IOM report supports this view wholeheartedly.

If the argument is framed in the correct manner, I think it is still very relevant.

 

That does not work logically. If a patient with PEM conforms to Oxford then findings in an Oxford population apply unless you have evidence that they do not when there is PEM. If you find ibuprofen works in an unselected population of people with joint pain then you can expect it to work in a patient with RA who has joint pain, unless you have a specific reason to doubt it.

As I understand it the IOM say that if one is wanting information on ME patients then studies should not use Oxford criteria because there are more precise criteria. But I am not aware that they have said that valid conclusions from Oxford criteria populations do not apply to ME patients that fit those criteria.

I agree that white's statement that PACE is not intended to apply to ME/CFS seems to rule it out but he says it does apply to CFS. Moreover, there is no particular reason to take White's word for how to interpret PACE!

I quite agree that one can raise all sorts of arguments about why these studies might not apply reliably to a more narrowly defined population but people do seem to be raising upside down arguments when it comes to probability and set theory here.

 

The arguments about diagnostic criteria used in UK/Dutch CBT or GET studies is a distraction/aside when such studies have not provided any objective evidence of improvements.

 

This is the quote in full......

That's how he is defining CFS in his own words for the PACE trial.....the definition is.....defined simply as a principal complaint of fatigue that is disabling

Hence fatigue need be the ONLY symptom.

 

OK I agree.

I didn't realise (IOM) ME was a subset of Oxford but it seems it is.

 

I don't understand this....

Humans are a large group.

Men, women and children are subsets.

Are you saying that because men, women and children are part of the human population then all attributes of humans apply or should be assumed to apply because they are all human?

This seems wrong to me.....

I do understand the need to put best arguments forward but .... Like @BurnA says, I'm reluctant to let go arguments before we start.

 

I think the problem from my understanding is if you want to study humans you can have millions of subjects in your study but if you have no requirement to include children in the group your results cannot be applied to children.

The maximum amount of people who could respond yes to "do you have MS" in the UK is around 100,000. The maximum amount of people who could respond yes to "do you experience fatigue" is 65 million.

Its just not sensible to say, "oh well we know at least that within that 65 million some people will have MS so that cohort of 65 million is acceptable".

 

Surely this comes down to the definition of the group that the NICE guidelines are being written for.

If the target group is defined by ICC (say) and the prevalence of ICC was shown to be 0.4% (I am only guestimating these numbers for the sake of the argument) while the prevalence of Oxford was 2% then it could be argued that, even if ICC is a subset of Oxford (which may not be completely true) ICC can only make up 20% of an Oxford defined cohort. Hence Oxford studies have limited relevance to the ICC group for which the guidelines are being written.

It is my understanding that this is the logic behind the AHRQ's downgrading of Oxford based studies. So IF the NICE guidelines are being written for an ICC cohort, similar arguments should apply. This is not saying that the Oxford studies are bad studies because they used Oxford, just that they are not that relevant to ICC.

 

Analogy

If we had a scenario where migraines were just dumped under the term headaches. Then some headache researchers took a cohirt of headache and migraine sufferers to assess paracetamol and mindfulness as treatments and 70% found benefit that could then theoretically form the basis of the headache treatments in uk. But if actually the research was only 10% people with migraines and of them most of them actually didn't respond and those who did has very mildest migraines then you can see how utterly failed most migraine sufferers were by both the lumping in with headaches and the failure to recognise not only their more serious illness but that actually if mindfulness and paracetamol were tested on a group of pure migraine sufferers the results would be poor and no one would form their management on such an inadequate approach. And migraines aren't even harmed by paracetamol and mindfulness

 

Also we shouldn't be framing this as ME is a subgroup of Oxford fatigue. ME is a distinct illness, with using strict criteria, a population size equivalent to MS. It's simply not ok to keep letting it be screwed over and not recognised because the uk can't bring itself to recognise it. For ME we know CBT/get don't help many and can make worse, the narrative, because of Oxford research has instead become we have moderately effective, safe treatment.

 

No, simply that an attribute found in a population of humans is likely to be found in another example of a human. Note, also that in the case of responses to treatments we should have control conditions and demographic data on known subsets like men and women, so we can refine our predicted probability of response quite a bit further.

 

The problem is that dealing with NICE guidance is not about PACE. It's about a bunch of studies, some of which show objective improvements. The ones which show objective improvement use Oxford and have indications that patients were not physically disabled at baseline. How do we argue that those trials are not relevant to ME/CFS, except by arguing that the criteria used did not recruit ME/CFS patients?

I disagree completely. Oxford is the only criteria which requires that fatigue be the principal symptom, and that it must include mental fatigue. At best the Oxford criteria overlap partially with ME criteria, though I think even that is doubtful. But Oxford categorically does not fully encompass any other criteria.

 

I was basing my assumption on this diagnostic algorithm.

It starts with patient presenting with profound fatigue.

I am open to correction.