• Welcome to Phoenix Rising!

    Created in 2008, Phoenix Rising is the largest and oldest forum dedicated to furthering the understanding of and finding treatments for complex chronic illnesses such as chronic fatigue syndrome (ME/CFS), fibromyalgia (FM), long COVID, postural orthostatic tachycardia syndrome (POTS), mast cell activation syndrome (MCAS), and allied diseases.

    To become a member, simply click the Register button at the top right.

Article: A Hitch in its Step: PACE Trial Indicates CBT/GET No Cure For CFS, 60% of Patients Show No

Here's part of a letter that I've sent to a UK patient charity...
I'm posting it on here in case anyone wants to make use of any of it...
If anyone wants to, then please feel free to cut and paste or use it in any way...
I haven't studied the paper very intensely yet, so there may be important stuff that i've either missed out, or that i've slightly misinterpreted...
But I'm pretty sure that i've got the details accurate.
If anyone notices any mistakes, then please could you let me know so I can correct them.
Or if it could be improved on, then please comment or rewrite.
I know it could be better written but my brain isn't working well this month...

I've focused on questioning the validity and interpretation of the data, more than the validity of the methodology and patient selection criteria etc.

____________________________________________


I am very worried about the results of the PACE Trial, and I am worried that the spin put on the results is going to cause immense damage to our community, possibly setting back the understanding of our disease, and it's representation in the media, by many years.

I believe that there are many ways in which the trial was manipulated in order to get the results which the authors desired. Such as using unofficial diagnostic criteria, and abandoning the use of actometers (the only objective measurement being used) during the trial.

Also, a lot of spin has been put on the results by the authors. Only 16 to 60% of the patients were helped by CBT or GET in the trial, depending on which figures you look at in the paper and how they are interpreted. And of those patients helped, they only experienced about a 10% reduction in fatigue levels. Personally, I would call this a 'small' reduction, and not a 'moderate' reduction, as the authors call it. I would describe a 'moderate reduction' as something around 35 to 65%, not 10%.

I've looked up 'moderate' in the dictionary, and it means 'not extreme'. So a 'moderate reduction in fatigue levels' means 'not an extreme reduction in fatigue levels'. I don't think that this accurately reflects the very small reductions that were recorded in the study. I think that 'small reductions in levels of fatigue' would be a more accurate description.

It also seems that the statistical data were skewed to a particular outcome in the first place. Because so many of the original fatigue scores were originally very high, as recorded on the Chalder scale, they only had one direction in which to go, which was downwards. Any of the participants who originally recorded very high fatigue scores, or the highest score, and then later experienced higher fatigue levels would not have been able to go above the highest score of 33. For this reason, the statistics were not a fair portrayal of the actual events as they were projected to go downwards even if only random deviation was recorded in the statistics. If a patient can't record a higher level of fatigue during the study, then this skews the wider results because the general fatigue levels are statistically projected to trend downwards anyway. As some patients improve, and some patients get worse, only the improving scores can be recorded because the worst (highest) scores are at maximum levels already. The use of the Chalder scale is unhelpful in this way, as the baseline average scores were so high (e.g. 28.2 out of 33 for GET.) Thus, I think that the results of the trial may not be a fair representation of events. Without access to the raw data, it is impossible to see how many patients recorded near-maximum fatigue levels throughout the study. Also, if patients recorded a maximum baseline score of 33, then it is impossible to monitor or record any deterioration in those patients.

I also think it is extremely damaging for the ME patient community to be told that 'pacing' is not helpful. This is untrue anyway. The study used a hitherto unknown, and unheard of, therapy named Adaptive Pacing Therapy (APT), especially formulated for the study, which clearly had little effect in the study. But APT is not at all similar to the intuitive form of pacing that patients use. APT is prescriptive, systematic and formulaic, and is clearly unhelpful, unlike 'pacing' which is intuitive and flexible and is very helpful for patients. Patients in the APT group on the study were also told that they would never get better, but had to accept their illness as it was, whereas patients receiving CBT and GET were told that they would get better using these therapies. This surely must have skewed the results, as patients with higher expectations, hope and motivation due to being told that their therapy would lead to an improvement in their health, may well be motivated to record better results when answering questions about their fatigue levels, when compared to a group who were told that they would never get better.

The fact that some patients could walk an extra 45 meters in a 6 minute time period (as compared to the Specialist Medical Care [SMC] only group), but still recorded high levels of fatigue (20.6 out of 33 on the Chalder scale after 52 weeks) does not seem like a successful outcome of the study to me. These results seem to tell us very little except that ME is clearly not a psychological disease, and cannot be effectively treated with psychological interventions. The fact that no patients were cured by this trial is very telling about the nature of the therapies on offer to our community, and about the nature of the disease. The poor results also tell us that the symptoms of ME are not caused by mental or physical deconditioning, or maladaptive behavioural patterns or maladaptive cognition, because all of the best psychological therapies on offer failed to actually cure any patients. The CBT and GET in the trial probably did no more for ME patients than they would to motivate any other section of the population.

In the GET group, 93% of patients experienced 'non serious' adverse events, and 9% experienced 'serious' adverse events. Also, 9% of the CBT group and 6% of the GET group recorded 'serious deterioration' in the trial. These results suggest that these interventions are not at all 'harmless' as they have been presented by the authors. The therapies cannot be described as 'harmless', like the authors are portraying them, when so many adverse events and 'serious deteriorations' have been recorded. The CBT group also recorded very similar adverse events as the GET group. We can imagine how these therapies would cause far more damage to patients, with far worse consequences, when administered outside the close controls of a very expensive clinical trial, carefully overseen by the 'best' experts in the field.

In the CGI figures in table 5 of the published paper, 58% to 60% of patients recorded either negative or minimum change in the CBT and GET groups. Conveniently for the authors, no separate figures were recorded for total negative change (which includes the minimal negative change data). Even so, these figures do not look either impressive or safe. For both the GET and the CBT groups, only 16% extra patients recorded 'positive change', as compared to the SMC-alone group (i.e. only 16% extra patients in the CBT and GET groups were helped as compared with the SMC-only group.)


______________________________________________
 
Nice one, Bob.

Patients on the study using this therapy were also told that they would never get better, but had to accept their illness as it was, whereas patients receiving CBT and GET were told that they would get better using these therapies. This surely must have skewed the results, as patients with higher expectations and motivations when using a motivational therapy would surely have had more motivation and hope which may have led them to record better results when answering questions about their fatigue levels.

This is particularly troubling, and seems pretty unethical to me, like they deliberately tried to discourage the APT patients from even trying. That is certainly not my understanding of pacing as patients use it, including me.

Only 16% extra patients recorded 'positive change', as compared to the SMC-alone group, for both the GET and the CBT groups (i.e. only 16% extra patients in the CBT and GET groups were helped as compared with the SMC-only group.)

Do you mean something like "only a net of 16% of patients reported improvement"?

This is much the same number of net responders as reported in the most recent Cochrane review of CBT, where 40% of patients in the CBT treatment arm and 26% of the non-treatment, "usual care" control group reported 'improvement' (= a net of 14% of patients reporting improvement that may be attributable to CBT).

So their very well funded, large scale, top quality study, run by world experts (who believe passionately in the model being tested), and using highly skilled staff, with lots of oversight - ie the best of possible circumstances for a success for their model - was still only able to improve the numbers of patients reporting improvement by 2% over the average CBT study. They are also coming off a very low base as well, so the actual real world gains for patients are non-existent, and if follow up results in previous CBT studies are anything to go by then even this minimal effect is not going to be sustained and will probably fade to statistical insignificance.

And this is before we factor in all the serious methodological problems, like patient selection, the definition of APT, and outcome measures.

Pretty unimpressive numbers.



Does anybody know what the actual closing date for submissions are for letters to the Lancet about the PACE study? I know it is two weeks after initial publication of the paper, but is that online or print version?
 
Does anybody know what the actual closing date for submissions are for letters to the Lancet about the PACE study? I know it is two weeks after initial publication of the paper, but is that online or print version?
I'm fairly sure that it would be the print edition. It costs a lot to subscribe to a journal so they'd have to allow them time to reply. So it's two weeks from whenever it is published.

I hope you write, Sean, as I think you could write a good letter.
 
Oh damn, I forgot to include a mention of the recently published sister-study, the FINE trial, which showed that GET has no benefit for severely ill ME/CFS patients who are house-bound and bed-bound. (It's funny how we never hear anything about that study, isn't it?!)

Thanks for the feedback, folks.

That is an excellent letter, Bob. I know I'm probably just whistling Dixie by suggesting this (American expression - means dreaming about a lost cause) but have you sent it to the UK media as well?

Thanks urban, no I haven't yet... I think it needs refining and would need to be changed for a lay audience... I can't do any more brain work today, so that'll have to wait... I think I might send Cort's article to some journalists, such as The Guardian's health writer, to attempt to educate them on how to do journalism!


This is particularly troubling, and seems pretty unethical to me, like they deliberately tried to discourage the APT patients from even trying. That is certainly not my understanding of pacing as patients use it, including me.

Yes, we're surely not going to get objective data if the different groups were told to expect different outcomes, especially when the entire study is based on psychological interventions. The placebo effect should have been accounted for when formulating the study!

Do you mean something like "only a net of 16% of patients reported improvement"?

Yes, exactly that, for both CBT and GET.

This was the "Participant-rated clinical global impression of change in overall health"...

Positive change reported:
SMC alone 25%
GET+SMC 41%
CBT+SMC 41%
 
Patients in the APT group on the study were also told that they would never get better, but had to accept their illness as it was, whereas patients receiving CBT and GET were told that they would get better using these therapies. This surely must have skewed the results, as patients with higher expectations, hope and motivation due to being told that their therapy would lead to an improvement in their health, may well be motivated to record better results when answering questions about their fatigue levels, when compared to a group who were told that they would never get better.

I think there are papers showing that patients generally like to tell the therapist whatever will make the therapist happiest. If one group of patients is told that a treatment will help htem get better, and the therapist then puts a lot of effort in to trying to make this so, it seems rude not to ssay you've gotten a bit better!

I've already mentioned that this could skew drop-outs too. Those who illness meant they were getting worse/not improving would be more likely to leave CBT or GET, where as those who were naturally improving might see the restrictive version of pacing being promoted as flawed and drop out.

The way they've presented and promoted their findings really does seem dishonest, and that makes it impossible to give them the benefit of the doubt for anything. Which is a pain.
 
Thanks Cort. Can you tell us more about the pairing of the treatments? Did the same patient group get to try out more than one treatment? For some reason I got the impression from listening to an audio I found that those that did the GET also did the CBT.
 
It does seem like their version of CBT includes elements of GET and vice versa.

I did not mean to suggest that at all. I meant that if only the ones that did the GET got the CBT and only the ones that got the adaptive pacing got the specialist treatment then basically there were 2 groups of 2 treatments. I am not sure if I got that right - but if it was that way then CBT as positive thinking technique gives unfair advantage to GET.
 
I did not mean to suggest that at all. I meant that if only the ones that did the GET got the CBT and only the ones that got the adaptive pacing got the specialist treatment then basically there were 2 groups of 2 treatments. I am not sure if I got that right - but if it was that way then CBT as positive thinking technique gives unfair advantage to GET.

Hi sparkle...
A very big welcome to the forum to you!

As far as I am aware, the groups were not swapped at any time.
Quite often in studies like this, they do swap the groups over, but I haven't heard anything about them doing that in this study.
All the patients received the Specialist Medical Care, and then the GET, CBT and APT groups received their therapy on top of the SMC.
 
PACE Trial letters ideally should be in on or before Friday March 4th

Somebody* who wrote to the Lancet to ask about the deadline got the following
response:
========

I asked the Lancet editors what the closing date is, and their response was:
----------
The two-week deadline for submission of Correspondence refers to the date
the relevant article was first published, so in this case the date of online
publication. You should therefore submit your letter as soon as possible.

We will still consider submissions received after this deadline, but we may
not be able to give them priority when compared to letters received within
the allotted timeframe.
----------
It was published on 18 Feb, so by my calculations this makes the deadline
the 4th Mar, which is tomorrow!
========

Submissions go through:
http://ees.elsevier.com/thelancet/

Remember 250 (or fewer) words
5 (or fewer) refs, 1 of which would normally be the paper being responded to.

Please tell anyone who might be interested.

* will add name if person wants