Member comments wanted: Second section (Who the guideline is for) of the NICE ME/CFS guidelines draft scope

The consultation for the draft scope closes 26 July 2018 at 5pm. Therefore, we need members to look at each section of the draft scope and provide feedback, in order to inform any comment that S4ME may make on the draft. So please read the below and comment away!

(Full document can be found here, NICE webpage on the process is here, we did have an initial thread on the draft scope here but these individual threads are to capture comments on the individual sections of the draft.

Thread for the first part of the scope (Why the guideline is needed) can be found here, and for the third part (What the guideline will cover) here.)

NB: There isn't actually much to this one.

2. Who the guideline is for

This guideline is for:

• All health and social care professionals including those working or providing input in to educational and occupational health services
  
• Commissioners
• People with suspected or diagnosed ME/CFS, their families and carers and the public

It may also be relevant for:

• education services
• occupational health services
• voluntary sector organisations

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive.

Equality considerations

NICE has carried out an equality impact assessment during scoping. The assessment:

• lists equality issues identified, and how they have been addressed
• explains why any groups are excluded from the scope.

The guideline will look at inequalities relating to ethnic origin.

ETA: adding links in the text.
ETA2: Add text and link back to original thread on the draft coming out.

 

their equality impact statement doesn't seem to go into much detail .about the people with m e who cannot travel or even leave their homes .

 

They emphasize that this will be based on the evidence and they acknowledge the questions on CBT and GET. But they fail to acknowledge the problems with the use of non-specific definitions such as Oxford and what they intend to do about that.

 

There was quite a lot of discussion about definitions at the stakeholder meeting. The key point for me is that diagnostic criteria are not in fact of great relevance to clinical care. Guidelines for rheumatoid arthritis would make no mention of diagnostic criteria for RA. What matter with clinical care are specific component problems that point in specific prognostic or treatment directions. Since we do not have evidence for any specific treatment being any use for any subgroup I am not sure how much it matters how wide the net is cast.

 

But I wonder if it matters in terms of preventing harms. I wonder if people with chronic fatigue who don't have PEM are at such serious risk of harm from GET (and GET-promoting CBT) as people with PEM (and not necessarily chronic fatigue, since not all PWME have that).

 

We don't know. I am sceptical of the idea that GET might be useful for 'other' sorts of fatigue. It was after all designed specifically for people with fear of exercise, presumably because of PEM. My concern is that you end up with this business of 'some people benefit from this and some from that' which is the slippery slope to keeping GET in the guideline. As far as I am concerned GET has no business anywhere in the guideline.

 

So am I, but my concern is that by diluting the NICE guidelines to apply to 'chronic fatigue' (the Oxford definition), we might dilute the population to include a lot of people who might not be harmed by GET. (OTOH, there's enough anecdotal evidence of harm so that maybe that doesn't matter).

But PACE didn't select patients on their fear of exercise (as far as I'm aware, no trial has done so) - it selected on the basis of having chronic fatigue.

PACE and lots of the other studies used Oxford and don't show credible evidence of benefit (because they used the same open-label design and subjective primary measures). CBT/GET should be discontinued on that basis.

I do think we might need to flag up problems with the lax criteria, though - future studies shouldn't be lumping together all patients with unexplained fatigue when there's a key feature like PEM.

 

I'm not sure if this is the correct place to post this but it is a question I have raised a couple of times (on the WHO CD11 threads). The UK divide ME/CFS patients into Mild, Moderate, and Severe. Given that it is quite possible that many GPs will have probably never seen a real severe patient how exactly are they supposed to determine which 'category' patients fit into?

If they are going to suggest 'treatment' based on this differentiation it should be made clear how this distinction is made. Similarly, it would help chart any progression of the disease, either up or down, (this might also be of use in getting the DWP to understand that people can/do get worse).

eta:
according to this review there are actually 4 classifications the fourth being very severe (but this doesn't seem to show up on any NICE or NHS literature(?)) and is not on WHO ICD coding.
http://www.meresearch.org.uk/our-research/completed-studies/severe-cfs-review/

 

NICE has nothing to do with studies, @Sasha. It is about providing guidelines for clinical care. You cannot have a situation where someone who does not quite fit criteria is told - sorry go home, we have no guidelines for you. Clinical care is not about putting people in diagnostic pigeonholes. It is about trying to judge the best way to deal with whatever problems a person may have.

I think it is important not to make life difficult for this committee and to encourage them to focus on the only reason why they are reviewing the guidelines and the one that matters - to re-assess whether CBT and GET should be mentioned at all. Whenever committees find things complicated they go for a compromise. I think they should be encouraged to see things as simple and go for the right answer.

 

I appreciate your point. But presumably evidence-based clinical guidelines for a particular disease should be based on an evidence base of good science in people that actually have that disease. That PACE and related studies are bad science is a critical point. But equally important is what AHRQ found - that the CBT and GET studies included patients with other conditions. Even without reassessing the quality of these studies, AHRQ downgraded their recommendations for CBT and GET simply because the studies included people other conditions and thus lacked external validity for patients with ME.

I dont think its either one or the other. The combination of bad science and also lack of external validity is powerful. It not only undermines CBT and GET as treatment recommendations but underscores that diagnostic accuracy is essential

 

That would be an illogical thing to do and I am afraid that it would not surprise me if a committee of that sort got it wrong. If CBT and GET were steroids and studies were conducted on 'any old autoimmune rheumatic disease' rather than lupus, it would be reasonable to say that if the steroids showed a useful effect the prima facie case would be that they would be useful for lupus patients. We do this all the time. We have trials showing painkillers help for joint pain so we assume they work for elbow pain. Trials are always done on rather wider categories than those any one person might be sub-setted into. That may weaken the predictive power of the evidence somewhat but we can never expect 100% predictive power in medicine.

So the fact that studies have been done on a wider disease set is not a reason to reject them. For PACE there is a more complex reason to reject the cohort because ME with PEM patients may have been seriously under-represented and there is an a priori reason to think that GET would be unhelpful specifically in the context of PEM.

Moreover, as indicated to Sasha, none of this is relevant to the scope of the guideline as advice on practical clinical care. That has to cover a pragmatically defined group of people with a clinical problem. Maybe there should be a PEM guideline but although PEM seems to be an indicator of a particular pattern of illness, so far we have no clear indication whether that is in fact a single or watertight disease category and crucially we do not actually have any evidence that it alters what treatment to recommend. I think it would be useful to flag up PEM in the guidelines as something to discuss in the context of coping strategy and it would be reasonable to mention that patients have reported that trying to push at exercise may be counterproductive but as for most of medicine that is treating a problem on its merits rather than trying to squeeze people into pigeon holes and working from there.

 

The problem with accounts like this is that they do not provide us with useful evidence.

The suggestion is that the problem is not fatigue but 'flat batteries'. But nobody knows what that means. It is an interpretation rather than a symptom or physical fact.

'too late to be aware of the cumulatively damaging effects of the episodes'
The problem here is that we have no evidence for cumulative damage. People get worse after crashes and the more crashes they have the worse they may get but that would be expected if crashes are a symptom of the problem getting worse for whatever reason. It is not possible to draw any conclusion of cause and effect.

It is reasonable to hope that the guidelines would include comments about patients having crashes and finding they do better if they eep to a low even level of exertion. But speculations about cause and effect will simply encourage the committee to think they are being presented with 'unhelpful beliefs'.

 

I agree with everything Edwards has said.

1) Don't complicate the simple. Keep the fight on CBT and GET.

2) I think the entire debate regarding whether some forms of CBT and GET can help some patients who suffer from fatigue is a huge mistake. I think Norway is a good example, where there has been a lot of great discussion. But sadly, a lot of the debate has suffered from the idea that the results of the PACE trial could be explained by the fact that some patients, who do not suffer from PEM, might actually benefit from CBT and GET, and that this could explain the minor increase in self reported fatigue and function. The discussion therefore inevitably becomes a discussion of diagnostic criteria, but also offers a needless concession to the BPS crowd, that CBT and GET could be helpful to ''some''.

Once again, do not complicate the simple. The entire problem with PACE and the similar studies is that the study is not blinded and relies on subjective outcome measures. Looking at the studies objective outcomes, it doesn't even look like people who don't fit the strict ME criteria get any improvement at all. There is no need to come up with a lot of different rationalizations about the results of the PACE trial, they are perfectly explained by lack of blinding and subjective outcome measures.

 

Just to clarify - the AHRQ evidence review was not done by an AHRQ "committee" but was commissioned by them and done by one of the evidence review practice centers in the US. Their 2014 review included Oxford studies like Cochrane did and concluded that CBT and GET were moderately effective. But in 2016, the evidence review practice center redid the analysis, eliminating Oxford to see if they would reach different conclusions. They did - they concluded that there was insufficient evidence to conclude GET efficacy and barely any for CBT. They also noted that studies using definitions that required hallmark criteria such as PEM were "blatantly missing."

I would think it would be highly relevant - PEM is the hallmark of the disease (even NICE seems to emphasize it) and there is significant evidence that the PEM is associated with impaired aerobic energy metabolism. If PEM is accepted as such, then basing treatment recommendations on studies in people who do not experience that symptom seems clinically irresponsible. Especially when the recommendation is to go exercise. We would not base treatment recommendations for people with asthma on studies in people who do not have asthma.

Edit:
In the US, some medical guidelines have resisted removing CBT and GET on the claim that it helps "some" people with "ME/CFS." But that claim rests on the idea in Oxford and Fukuda CFS that all medically unexplained chronic fatigue should be studied and treated as the same thing. Medically there's no reason to expect that all causes of chronic fatigue could be treated in the same way. CBT and GET may help some people who have chronic fatigue but as AHRQ noted, there is no evidence that it helps people who have ME.

 

OK it sounds as if they used something a bit like Cochrane - which presumably had a committee. And like Cochrane it looks as if they don't get the bit that matters - that there is no usable evidence for efficacy of either treatment for anyone in any of these studies. As @Rick Sanchez puts neatly, the mistake is to think that the problem is that there is evidence of these treatments working for non PEM patients that is giving the confusing results. There isn't even evidence for that. Unless the NICE people get that they will fluff things.

PEM being a hallmark of a disease is one way of looking at things as we are now, but as I have indicated, disease pigeonholes are not what makes clinical policy. Clinical policy is based on the real life need to make complex assessment of probabilities for each of a spectrum of people that evidence from studies on stereotyped cases is relevant to them. Disease names are often not the important indicators. (And I am not convinced that CPET data tell us about PEM, as has been discussed in another thread.) Again the bottom line is that nobody is suggesting any treatment recommendations derived from studies on people without PEM because there is no evidence on which to base these recommendations. We don't actually have a problem in that regard. We have a problem with a lot of dim evidence-assessors not understanding the basic rules of evidence reliability. They need to have their noses pointed in the direction of the basic error, not encouraged to be distracted by a possibility that ignores that basic error.

 

Yep. Complexity breeds confusion which leads to cock-up.

 

The diagnostic criteria probably do matter in the sense that patients who are less impaired are more easily manipulated into reporting improved health. That doesn't mean they never had ME or CFS to begin with or that they are genuinely improving.

In case it is not clear, I fully agree with the root problem being methdology that doesn't sufficiently control for bias, rather than diagnostic criteria.