Collection of evidence for the efficacy of CBT/GET

[DMissa]

Hi all, it came up in another thread that it would be good to have a one-stop shop of all papers or evidence examining the efficacy or risks of CBT/GET for people with ME or LC (LC-ME preferred but my guess is that no relevant studies taking into account PEM specifically will have been done yet in the LC sphere). If anyone is able to help with this it would be great (both evidence in favour or against btw). I am working on an extremely short timeline for something confidential but hugely important.

 

[SNT Gatchaman]

There'll be secondary references in these.

CBT and graded exercise therapy studies have proven that ME/CFS and long COVID are physical diseases, yet no one is aware of that (2025, Frontiers in Human Neuroscience)

Graded exercise therapy for myalgic encephalomyelitis/chronic fatigue syndrome is not effective and unsafe. Re-analysis of a Cochrane review (2018, Health Psychology Open)

Rethinking the treatment of chronic fatigue syndrome—a reanalysis and evaluation of findings from a recent major trial of graded exercise and CBT (2018, BMC Psychology)

 

[forestglip]

The NICE guidelines may be useful, specifically this document:
[G] Evidence reviews for the non-
pharmacological management of ME/CFS

It has the detailed analyses of all the studies they looked at. The overall conclusion for CBT begins on page 372. And the conclusion for GET begins on page 383.

 

[Hutan]

There's this thread:
Evidence of GET (and exercise generally) being helpful or harmful in ME/CFS and related conditions

 

[Hutan]

From that thread:

I would point to the NICE Analysis of Evidence for the 2021 ME/CFS Guidelines, which found that there was no reliable evidence that graded exercise therapy was helpful. That finding resulted in the Guidelines recommending against graded exercise.
NICE guideline website: https://www.nice.org.uk/guidance/ng206

I would also point to the MAGENTA trial that found that graded exercise therapy and graded activity therapy were not helpful for children with ME/CFS.
Forum thread: Graded exercise therapy compared to activity management for paediatric [CFS/ME]: pragmatic randomized controlled trial, 2024, Gaunt, Crawley et al.

On harms:
Forum thread: Do graded activity therapies cause harm in chronic fatigue syndrome? (Kindlon, 2017)

We have extensive coverage of the Cochrane Larun et al review on the forum. Regardless of whether it has a publication date of 2019 or 2024, it only covers studies undertaken in 2014 or before. Unlike NICE in its 2021 Guideline Review, the Cochrane review fails to identify a litany of basic methodological issues that result in the studies it includes having a low level of reliability. It also does not pay adequate attention to the issue of harm.
Forum thread: Cochrane Review: 'Exercise therapy for chronic fatigue syndrome', Larun et al. - New version October 2019 and new date December 2024

The PACE trial is the biggest and most influential of these studies. We have extensive coverage of that trial on the forum too. A major problem in that trial and most others of exercise therapies for ME/CFS is the reliance on subjective outcomes in an unblinded study. The subjective outcomes are typically surveys asking the participant if they feel better and if symptoms and limitations have improved. The bias that that approach produces is made worse by the hyping of the preferred treatments as successful and the indoctrination of participants in the idea that they can get well if they put the work in (and so, if they do not get better, they have failed).
Forum library thread with links to resources: The PACE Trial
Forum thread on a series of videos explaining the problems: Video: The PACE trial: a short explanation, Graham McPhee

On harms, see our letter to Cochrane which spells out the evidence with references:
https://www.s4me.info/threads/s4me-...rcise-therapy-review.34973/page-2#post-521800

 

[Hutan]

I am working on an extremely short timeline for something confidential but hugely important.

It's a bit of a shame that something hugely important both has a short timeline, and doesn't allow for review by the people who will be affected by the analysis.

'Nothing about us without us'...

Good that it is in your hands though Daniel and that you are here.

 

[Hutan]

If you click through the threads tagged with GET or CBT, you will find some useful papers (click on tags at top left of this thread).

e.g.
"The idea of rehabilitation needs to be completely rethought for this disease" A qualitative analysis of patients' experiences, 2023, Hammer et al
That one is focused on Post Covid Syndrome and PEM

Evaluation of a survey exploring the experiences of [people...] participated in CBT and GET..., 2019, OXCATTS

 

[Utsikt]

@dave30th might have some examples on hand from his presentations?

 

[Jonathan Edwards]

My memory is that we have seen LC studies on rehab programs that do talk about PEM but argue somehow that you can plan around this.

There are masses of papers on this for ME/CFS or CFS, nearly all of which claim to show efficacy. So collecting them all together isn't a very helpful exercise unless the reader understands why none them are methodologically valid. The critique papers mentioned above go into this but largely for PACE, which comes closest to being meaningful but still falls short.

My expert witness submission to NICE explains the nature of the problem, also given in my JHP paper.

https://www.nice.org.uk/guidance/ng206/evidence/appendix-3-expert-testimonies-pdf-333546588760

Edwards J. PACE team response shows a disregard for the principles of science. J. Health Psychol. 2017;22:1155–1158. doi: 10.1177/1359105317700886.

You can guarantee that all studies done on LC will suffer the same validity problem but people who work in the field of therapist-delivered treatments simply do not understand why there is a problem - and sadly nor will most people being shown the evidence unless they are familiar with trial design.

 

[Utsikt]

SIPCOV found that their treatment for LC was ineffcetive, even though the authors and Chalder claim that it was. They did not reach the MCID threshold of 10, they got 9.2.

They claim to have measured PEM (the postexertional malaise (PEM) score from DePaul Symptom Questionnaire-2), but knowing the authors they don’t believe in real PEM. And there are plenty of issues with the questionnaire as discussed elsewhere.

There are some comments on the article page as well, e.g. from Vink.

https://www.s4me.info/threads/brief...et-al-2024-with-comment-from-t-chalder.41728/

 

[Utsikt]

MAGENTA is pretty clear cut:

CONCLUSIONS
There was no evidence that GET was more effective or cost-effective than AM in this setting, with very limited improvement in either study group evident by the 6-month or 12-month assessment points.

https://www.s4me.info/threads/grade...trolled-trial-2024-gaunt-crawley-et-al.37488/

 

[Trish]

On the often repeated claim that GET is only harmful if it's not supervised by experts, that has been shown to be untrue by this survey specifically only completed by people who were prescribed GET at specialist ME clinics in the UK after the 2007 guideline which recommended CBT/GET.
OxCATTS, 2019
And there's a good article on the same problem of false claims that GET is only harmful when done without expert supervision.
Kirke (2017)
We included both of these references in our Cochrane submission on harms.
https://www.s4me.info/threads/s4me-...rcise-therapy-review.34973/page-2#post-521800

I think this issue is important. It's used in the PACE GET manual, and I noticed it's in the Australian GP's submission to the guideline consultation. It's simply untrue. In fact a very large proportion of those reporting harm in surveys from GET did it on the advice of specialist clinics.

There are also large European and Australian surveys on CBT/GET. I hope someone else has the links.

 

[Trish]

"The idea of rehabilitation needs to be completely rethought for this disease" A qualitative analysis of patients' experiences, 2023, Hammer et al

 

[Utsikt]

The Norwegian ME Association has a page on GET and CBT with links to various surveys as well. The page is in Norwegian, but auto translate in the browser should be good enough:
https://www.me-foreningen.no/om-me/behandling-og-pleie/kognitiv-adferdsterapi-og-gradert-trening/

 

[Sasha]

Hi all, it came up in another thread that it would be good to have a one-stop shop of all papers or evidence examining the efficacy or risks of CBT/GET for people with ME or LC (LC-ME preferred but my guess is that no relevant studies taking into account PEM specifically will have been done yet in the LC sphere). If anyone is able to help with this it would be great (both evidence in favour or against btw). I am working on an extremely short timeline for something confidential but hugely important.

Given that you're working on an extremely short timeline, you're not going to have time to read and synthesise tons of papers if your aim is simply to summarise the evidence on efficacy and risks.

So on efficacy, I'd focus on the trial design issues: namely, that in open-label trials, all subjective outcomes suggesting success can simply go in the bin. I'd also explicitly address the defences that BPS proponents have put up against them, because they seem to be pervasive and to explain why these trials continue to get funded, get past ethics committees, and get published. They seem to be:

• not understanding why potential bias in this situation means that this is a fatal flaw;
• assuming that bias can only make a genuinely positive effect bigger, rather than making a null effect look real, or a harmful intervention look like a helpful one;
• thinking that if only badly done studies exist, we have to treat their results as though the studies weren't badly done;
• if a lot of 'scientists' say we should take open-label trials with subjective measures at face value even when it's illogical and other scientists disagree, we should accept it.

A few trials have had objective measures but PACE was by far the biggest and so is where to focus, but it would be handy if there was a meta-analysis of objective effects across trials. Does anyone know if there is one?

It's a pity we don't have a handy, concise summary of all this stuff for exactly this kind of occasion. It would be very handy for advocacy. Do we really not have such a thing?

 

[DMissa]

doesn't allow for review by the people who will be affected by the analysis.

I can't disclose what this is for, but there are definitely affected individuals closely involved. What their clinical pictures are and how aware of this topic they are is unknown to me, which is why I want to have the evidence base at hand if it becomes relevant.

There are masses of papers on this for ME/CFS or CFS, nearly all of which claim to show efficacy. So collecting them all together isn't a very helpful exercise unless the reader understands why none them are methodologically valid.

Correct, also why I have asked to be pointed towards details in particular.

I am sincerely thankful to everybody for your help. I know that these ordinarily small asks might be much more difficult for some of you than they would be for me. I only ask due to importance + timing.

Much love.

 

[Trish]

If you want an excellent summary of key problems with GET research organised under headings, Michiel Tack has done an excellent comment on the Cochrane review of exercise therapy.
https://www.cochranelibrary.com/cds....pub9/detailed-comment/en?messageId=266353165

 

[Sasha]

@DMissa, if you want rapid feedback on your draft, I'm sure there would be people willing to help here, and if you wanted it to be private you could get the feedback via a group PM, for example.

 

[Turtle]

The new documentary "Doctors with ME" by Anil van der Zee, might be helpful too. Marc Vink, GP, who also published about GET, becoming very severe because of GET.
You can find it at the end of the documentary.

And a psychiatrist an ass. psychiatrist explaining why CBT is not supposed to be used for physical illness, but for e.g. agoraphobia etc. (false beliefs)

 

[Utsikt]

thinking that if only badly done studies exist, we have to treat their results as though the studies weren't badly done;

I believe they say something more like «it’s our best so it has to be good enough».

@ME/CFS Skeptic has a rebuttal of that kind of reasoning here, from Flottorp’s response to NICE:
https://mecfsskeptic.com/a-rebuttal...-fatigue-syndrome-more-ideology-than-science/

And @Jonathan Edwards adresses how we might need entirely new trial methodology to test e.g. CBT in his statement to NICE:

Trials to date have failed to make use of methods that can significantly mitigate difficulties with bias from subjective outcomes, but it may be that for unblindable treatments new forms of trial methodology will be needed if useful assessment is to be achieved.

https://doctorswith.me/wp-content/uploads/NICE-2021-Expert-Testimony.pdf