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

The point is that the BPS and PACE crowd do not "cast a wide net" therefore potentially catching multiple groups of a syndrome, they instead deliberately attempt to be specific in recruitment yet apply their "findings" broadly. That's unscientific and deceptive and they do it deliberately knowing they can shape policy because in the past the policy makers didn't look beneath their claims.

If they were to use varying criteria as you explained sometimes happens in RA that may be different but they deliberately use the same specific criteria in EVERY study. That in itself doesn't contribute to the scientific understanding especially when their criteria are designed to carry a political outcome not to answer a broader scientific question of something THEY themselves declared a syndrome .

If its political its not scientific.

 

Because its a biochemical imbalance, especially if it lasts longer then 2 weeks. 14 days is acceptable, no more

 

That's ironic, right?

 

CCC and Fukuda require fatigue, but not as the explicit principal symptom. ICC requires fatigability but not fatigue. SEID requires fatigue, but primarily seems to use it as a lead-in to the PEM requirement.

I doubt anyone with PEM (or OI, or pain from ME) would consider it to be a secondary symptom, only to be mentioned in passing after a principal symptom of fatigue.

 

Or it may stink because its not legitimate.

 

Here a link to download the PDF:
http://www.ncf-net.org/library/Reporting of Harms.pdf

ETA : Is this in our library @Woolie ?

 

I wish it were

 

But the use of the Oxford criteria by the BPS crowd in trials is not to be inclusive of all criterias by using Oxford its being used as a deliberate attempt to specifically exclude the rest of us.

 

I mean, you're ironic.

The BPS folk really propagate that crap.

 

The DSM 5 is the bible that says over two weeks and its depression

 

If you feel better on the 16th day then you had Major Depressive Disorder for one day.

 

@Alvin, @large donner, just great!

 

Sometime in the past I would have been burnt as a heretic.

Sorry, off-topic...

 

Nitpicking point Jonathan that doesn't in any way negate any of your other comments but as I understand it the behavioural part is not just a outcome of changing beliefs but and integral part of the process. For example with a phobia, you may claim to have changed phobic thoughts but it requires the practice of behaviours that challenge the phobia to demonstrate and reinforce the change.

I often feel that the GET side of CBT/GET was less about addressing presumed deconditioning than as the behavioural part of CBT. Sadly (sic) for the theory, on some measures the PACE trial showed that GET+SMC was slightly more effective (sic again) than CBT+SMC.

 

Maybe they do but that has nothing to do with using Oxford criteria being scientifically valid.

I am doing my best to root for the ME community but I am being a bit devil's advocateish because I want NICE to be presented with really crisp arguments.

In July they will pick 4 patient members for the committee. There will be I think 10 professionals. There will not be a big psychiatric presence but most physicians who deal with ME in the UK think 'biopsychosocial' is a good word. So the task is to set the agenda for them. And the agenda is absolutely rigorous argument about why PACE and its predecessors are not reliable evidence for anything - except perhaps that CBT and GET do not work at two and a half years. We cannot use arguments like 'the BPS crowd deliberately fiddled things', even if they did. Whether or not Oxford criteria are appropriate is a matter of opinion. Bad trial methodology is not a matter of opinion. If professionals are presented with weak arguments they will close ranks within minutes and the game is over.

 

Very interesting, all about good strategy

 

I think there is a cross-communication error here, and will stick my neck out to suggest how.

I think Jonathan is entirely right with his comments about scientific validity. If I carry out a study on the performance of students in mathematics in state schools in the UK, that could well be a valid study. Anyone who is a teacher, however, knows that, as a generalization, boys have a different attitude to the subject than girls, and that an overall finding that, say, coursework acts as a motivator, need not necessarily be appropriate: in fact girls tend to benefit, boys tend to do worse. It doesn't mean that the study is scientifically wrong. It means that whoever tries to apply it needs to make sure that the results are applicable to the group in question.

Even if such a study showed good results for both boys and girls separately, there would still be many children for whom the change would be worse. It is why the lack of good data on harms, as emphasized by @Tom Kindlon , is so crucial.

That is why the US have dropped any studies that use the Oxford criteria: they have decided that they are inappropriate for their purposes.

So the blame lies with NICE for relying on results that cannot be generalized to a more specific group, and on funding bodies, like the MRC, who continue to fund such fuzzy studies.

I agree with Jonathan that, if we want to criticize the science, we have to be very careful in the way in which we phrase our criticisms. Those here who were involved with the various analyses carried out on the released PACE data will tell you just how very, very careful folk had to be to get the phrasing right, and to keep the criticisms tightly focused.

 

I am not an expert on this, but I have some thoughts on it. And although I understand your way of reasoning and what you say - and although, in principle, I could agree to that reasoning - I am not sure you are 100% correct.

It is scientifically relevant how you choose your cohort for your trial. In fact, as I see it, your cohort will be the basis for inference statistics; it is your "representative" group.

Of course, in principle you can choose any cohort and apply your statistics on your data. The question is which conclusions or deductions you can make from this. And this is a scientific question. With clinical trials, it's THE question.

If it is not clear what your cohort is - and using the Oxford criteria gives a cohort with a certain non-uniqueness; it's not really clear what an Oxford cohort represents because it seems to be a mix of everything - how can you justify your conclusions? You cannot generalize. You could say: For this special group of persons we showed... Maybe you could even say that for most persons exercise and optimising behavior is helpful. (I.e. an Oxford cohort possibly represents the general crowd better than the ME crowd.)

If I understood correctly the PACE authors - as nearly all clinical trial authors - applied inference. And I would say that's an error.

I am as bold as to say that persons like Karl Popper, who influenced how science is practiced today, would reject such conclusions as non-scientific, i.e. not mirroring reality adequately. You have to be careful which deductions or inductions you draw. Each step of your way must be correct/proven.

Edit: Ah, I see @Graham in fact said the same.

 

Thanks @ Graham. I've been mulling this over and struggling.

I see what @Jonathan Edwards is saying. From the sound of things using wide criteria is done elsewhere in research. It doesn't in itself necessarily cause problems.

The problems arise in the application - ignoring and failing do look for subgroups who may be harmed and putting great big red stickers saying inappropriate for this subgroup.

It brings to mind Peter White's comments once 're the fact they were never studying CFS, a!though elsewhere he said entirely different things. Perhaps it was a early defence: we carried out this trial and NICE didn't implement it correctly so it's their fault? Load of guff but I wouldn't put it past them.