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

@Jonathan Edwards

Do you think that the Oxford criteria has a better potential to pickup up people who fit multiple criterias or a small subset of people within a syndrome?

Surely if it only picks up a smaller subset of people within a syndrome a panel of doctors/interested parties should be presented with that to show that using recommendations from a trial on that subset is foolish as an overal guideline for a patient population.

There's no reason not to present every other argument alongside that but why would you leave that obvious statement out?

This argument has already worked with the CDC and the IOM alongside the other ones you are proposing.

No one is saying one has to say 'the BPS crowd deliberately fiddled things'.

 

I disagree with your position.

As far as I'm aware there was concern in the USA that Oxford fatigue studies shouldn't be assumed to apply to the more complex sick ME/CFS which is why they were droppped from recommedations. Not because the USA considered the studies themselves necessarily junk.

The NICE guidelines are for ME/CFS. No serious ME group considers Oxford CFS good enough criteria for ME. In the USA the minimal criteria is Fukuda, vastly more stringent than Oxford. Oxford research shouldn't be being applied to ME which has exertion abnormalities as it's core. Any fatiguing illness can meet Oxford, you just need to be tired and maybe have other symptoms. It might exclude recognised illness but the point make is I wouldn't accept for example that Oxford results could be applied to MS so why ME?

By definition people with ME struggle with exertion and usually have many symptoms (systemic disease) . Using criteria to study exercise that doesn't require people to struggle with exertion is just impossible to use as supporting evidence for those with ME dumped under CFS. We aren't talking about a tiny subgroup here but the illness ME itself which has been totally failed because of the uk lump approach. (I use ME separately because in uk CFS is so broad), In USA where they use me/CFS it is still not Oxford fatigue.

 

But it is quite clear what Oxford is - a mix of everything. And Popper would have to be happy with that as long as conclusions were applied to a mix of everything. If you study fruit that is going in to a fruit salad then you can apply your findings to fruit.

The key thing is that there are things TERRIBLY wrong with these studies. Appalling errors. Things old Karl would have pointed his finger at with a twinkle in his eye and said 'no way, José'.

 

I may be wrong but I am pretty sure the problem with Oxford is that it picks up everyone. Isn't that your understanding. I cannot quite get your point here.

 

I did it deliberately, to see who would be first to pick me up on it .

Possibly. I recall a long time back now, well over 10 years, when my wife was first diagnosed with ME by a specialist, who I was impressed with then, and much more impressed now I understand the issues, which we had no idea of then and could easily have harmed my wife.

Having done all the usual elimination tests beforehand, my wife and I then spoke with this specialist, who asked my wife lots of detail (cannot recall much of it), including how she coped with her ME. In the end he advised us there was nothing he could do better than what my wife was already doing. We were somewhat disappointed at the time, but knowing what we know now, we are very grateful for his advice.

 

Agreed. Chronic fatigue is a symptom in many diseases, and is probably better described as a general symptom of illness rather than indicative of a specific illness.

 

That's what I agree to.

But ME is not a mix of everything.

He really would have done that.

 

Nooooooo!!!!!

Oxford is designed to attempt to capture non neurological non immune symptom based patients and maximise potential for tired people in the same way that Crawley attempts to find truant or tired teenagers in her trials yet goes on to claim that her treatments can be applied to other groups.

Yes there may be some who get through the net with neuro and immune based symptoms but its not by design. That's the whole point.

Why on earth would the BPS school risk including people in their trials that would potentially show less effect from their treatments? They use Oxford to deliberately exclude what would potentially be the condition they don't want in their studies.

That's the whole point.

 

Yes but no serious RA group or RA physician considers 'joint pain' a good enough criterion for RA. But studies of joint pain are perfectly reasonably applied to everyone with joint pain, including people with RA. Until someone shows that RA patients are an exception.

Let me put it another way. None of the ten professionals on the guidelines committee are going to be impressed by this argument unless they have theoretical reasons for believing that there is a subset of people with 'true ME' for whom conclusions from a broader category would obviously be inappropriate. If we are lucky there will be professionals with such theoretical reasons but I wonder who they are going to be?

I suspect it has been much easier for the US people to abandon CBT and GET because there are no government funded health care professionals who will suffer. I think the UK is a different ball game.

There is no harm in pointing out that the Oxford criteria are wide and since they are largely used by psychiatrists to recruit they may end up recruiting largely people with mood problems. But the point I am trying to continue to make is that that has nothing to do with Oxford based studies being scientifically unsound. They may be excellent for patients attending psychiatric clinics - if they are done well enough.

 

No it isn't the point. If the criteria allow 'true ME' patients in then they can be applied to them. Whether they are designed for some scurrilous purpose or not has no bearing on the scientific validity of their application to people who fit them.

 

The truth is though that we have no need to criticize the Oxford criteria. The small trials "supporting" CBT prior to 2007 used Oxford, and were only able to produce subjective improvements: all objective data said it was useless. PACE was set up to be definitive and, to its credit, definitely showed that they were right - it can only influence subjective assessment a little, and does not produce any meaningful real improvement. PACE wipes out previous similar studies, so at a stroke the Oxford criteria become irrelevant to past studies.

Our concern has to be Esther Crawley's even looser application of Oxford criteria on children, and for that we need a tight, scientific argument.

 

There was at least one GET trial which showed CPET improvements. Of course, the patients selected for that study had normal CPET scores even before treatment. So it basically just showed that physically healthy people can benefit from some exercise.

 

I guess you are saying it is not the Oxford criteria per se that is flawed; but misuse of it. A 2lb club hammer is an excellent tool, but not typically part of a watchmaker's tool kit.

 

But surely, surely the criteria have to be specific enough to define the disease in question otherwise they are of no value at all as diagnostic criteria?

If you try putting the word pain into Oxford instead of fatigue we may see what a nonsense it is then, you'd include all diseases with pain, from whatever cause, lumping them all in together.

 

So if you were in the BPS crew and had two doors one for entry to your studies marked "fatigue door" and one for no entry how many people would you include with neuro immune and neurological symptoms via the fatigue door?

 

I would disagree because i can invent a new drug for coughing and study people who have asthma, smoke cigarettes and have colds. If this drug works for asthma sufferers and no one else and my population is coincidentally made of 20% asthmatics and no control then my trial will show poor effect, because it was poorly designed, all had the symptom but my trial produced garbage data, and based on it i would not use this drug at all when in fact it may be 100% effective if only tested on asthmatics and would represent a promising treatment. If i didn't ask correctly in my study why they all had a cough i would not be able to pick this apart, because i shoehorned 3 conditions into one trial. If i run another trial with the same criteria but this one happens to have 70% asthmatics, my new drug will do great and i will have two opposing trials done on the same faulty premise with differing results, both technically valid but with glaring differences based on chance.

All that said i agree with your approach in this case, picking the best arguments for the target audience is the best way to get positive results.

 

I do not have a scientific background, and am particularly grateful for this forum for the clear explanation of the multitude if affected systems and factors.
Being relatively new to ME, with little statistical background, some threads are more challenging than others.

My main concern with Oxford is not its all encompassing definition, ( which cleverly seems to have guaranteed a particular line in scientific study and publications) it is its " tailored" application.

If the cohort within trials was representative of the defined group, then the full spectrum of conditions for which it is a definition for should have been trial participants. It is clear that this was not the case , and so how can results be generalised to the whole defined population ? Is this not a basic cock up?

Or perhaps my brain is more fried than i thought....

 

Oxford excludes "proven organic brain disease", so does not actually exclude PwME with neurological symptoms.

The short paper introducing the Oxford measure is here, the criteria are on the second page,

www.ncbi.nlm.nih.gov/pmc/articles/PMC1293107/

 

@Jonathan Edwards

I really think its a risky strategy not to point out how bias the Oxford criteria can be because if as to date so called high ranking policy makers haven't grapsed the flaws of unblinded subjective end point trials they may well go along with the original Lancet claim that 60+% of people improved and 22% recovered and the BPS crowd can then go on to claim that the GET and CBT strategies just need to be refined and rolled out more specifically to ALL OF US.

I'm just bemused that as we know multiple agencies in the US have dropped Oxford criteria explaining in detail the problems and as a result have removed PACE treatments from their advice that we think we shouldn't use Oxford as one of the reasons to achieve the same.

 

So yes, if the people enrolled into a trial includes PwME, but also people with depression, anxiety etc as their primary (or significant) condition, then it would be OK if the outcomes discriminated accurately which outcomes applied to people with which conditions, and the authors' analysis then applied validly to arrive at high integrity results. But of course they don't. The scurrilous bit is the results being interpreted as if everyone had ME. I suspect PW may originally have let something slip when he said they were only investigating chronic fatigue; maybe in PACE's embryonic stages there really was the thought to investigate chronic fatigue, soon lapsing into investigating CFS, ME, etc, maybe one or two people realising what a political money-saving coup it could be for them.