Chronic fatigue syndrome-Handbook of Clinical Neurology, Vol. 106

[Esther12]

Hopefully not that important, but I thought I'd post the link:

https://listserv.nodak.edu/cgi-bin/wa.exe?A2=ind1205d&L=co-cure&F=&S=&P=12748

Ta to Tate Mitchell for posting this, and Tom Kindlon for linking to it on twitter.

They don't do that good a job of pointing out a lot of the problems in the papers they cite, so it reads more like a contest to see who can get the most papers published than a careful examination of the evidence. It reads as if they've just accepted the analysis given in most of the papers, rather than looked to see if these interpretations of the evidence are reasonable... so be prepared for some annoying stuff.

Also, they say this:

Of all treatment approaches, cognitive-behavioral therapy (CBT) and
graded exercise are best supported by randomized, double-blind trials,
as well as meta-analyses of these trials.

Can anyone think of a single double-blind trial for CBT or GET? I don't see how such a thing could be done. Or single blind! Which is a bit of a problem when people are relying upon subjective questionnaire scores as outcome measures, particulalrly when these have been shown not to correlated well with objective measures of disability.

 

[Mark]

I can only guess that they would claim that when the researchers are allocating participants to the various groups, and when they're administering their questionnaires before and after the therapy, they don't know which were getting CBT/GET and which were getting a made-up therapy - and they don't tell the patients the name of the therapy.

I guess you could make the case that patients who didn't know what CBT or GET was might not be able to tell the difference between that and another equally useless made-up therapy...but it's still really stretching it to call it 'double-blind'. And this would at least assume that the 'control' groups would have to be getting some other kind of counselling and exercise therapies, in which case there's certainly no guarantee that those other therapies represent any kind of neutral 'placebo' type of intervention.

 

[Esther12]

Trying to do double blind trials on CBT or GET seems so convoluted and unlikely to work that I would be surprised if anyone had ever bothered. The idea of it is just absurd. But here we have it being claimed that it's been done repeatedly, with positive results and meta-analyses. It would be interesting to know what they meant.

 

[Esther12]

Oh... just saw URS M. NATER was the corresponding author... a particularly rubbish CFS quack imo. At least White and Wessely usually seem brighter than average. Nater's work seems to be consistently dire, as well as misguided. I'm now not surprised that they're claiming CBT/GET have been shown to be effective in double blind trials, or that they failed to spot/report the problems with the studies cited.

 

[alex3619]

"Taken together, these studies indicate that CFS is associated with
maladaptive illness beliefs and coping strategies, potentially
resulting in inadequate treatment efforts and regulatory adaptation to
challenge, with the consequence of persistent fatigue and other
related symptoms." - end of cbt section, my bolding

If you start with the presumption that CFS is a dysfunctional belief system this might make sense. If however you start with a presumption that CFS or ME is a biophysical disorder, then its nonsense. This is petitio principii or begging the question, a logical fallacy. Its entirely plausible that the belief systems are positively adaptive and good coping strategy - but to draw that potential conclusion you have to discount the idea that CFS is from dysfunctional beliefs. So depending on what you presume to start with, its either a good assertion or unfounded. This appears to me to be the hallmark of begging the question. Behind the fallacy is the hypothesis, completely unvalidated, that CFS is due to dysfunctional beliefs.

I see this kind of problem going back to Charcot in the nineteenth century. Irrational conclusions abound in this literature.

Now if they then went on to assert that some "maladaptive beliefs" justify the idea of a dysfunctional belief system as a cause of CFS, then you have a complete circle, the logical fallacy of circular reasoning.

Bye, Alex

 

[Don Quichotte]

The ideas of Charcot were questioned even then.
The reason they became so widespread in their use, is that they are so convenient.
And they still are...

when people fail (and a physician that doesn't know what his patient has and has no clue how to treat him fails in many ways) they tend to find an external reason for their failure (it's not that I don't know how to dance, the floor is uneven...) and when so many others physicians fail with the same patient, then it is obviously the patient's fault. (if no one, including the best dancers in the world can dance on this floor then there is clearly a problem with this floor...).

It is very easy to fall into the trap of (superficial) psychological explanations for the behavior of others (or even our own behavior). In fact, it is quite normal to try and understand the behavior of other people, what their motivations are etc.

And yes, a double blind trial of CBT is ridiculous. It's like a double blind trial of placebo.

 

[Don Quichotte]

It reminds me that when I was in medical school, I did a small study on a non-pharmacological approach for pain control in labor.
One of the gynecologists who was very skeptical initially, was quite impressed when (what he thought was an "hysterical" woman) had a wonderful childbirth experience and was very calm and happy throughout the entire process.
" I have to admit that there is something in what you are doing that really works" he said to me later, " but" he added, " how do you know that it is not just a placebo effect?"

For a moments this question really puzzled me, because how could I know that it wasn't placbo?
The term placebo was being used as a background noise that you have to eliminate in order to know how good a certain treatment really is.
But, then I thought about the original meaning of placebo=to please.
And this was exactly what I did, I created a non-judgmental calm, reassuring pleasant and warm child-birth environment, instead of the cold and detached medicalized environment that was prevalent then in large teaching hospitals delivery rooms.
And of course it worked, why wouldn't it?

From that moment on, I started to use " placebo" for each and everyone of my patients, as part of their medical treatment. I tried to create a non-judgmental warm, reassuring and pleasant environment (and was very fortunate to find a place to work in which shared my approach and preferred this position to one I was offered in a large prestigious teaching hospital ).

Structured, " one size fits all" CBT is in my humble opinion the exact opposite of that.
I believe that the reason it does work to some extent is that good and experienced therapists do not follow strict guidelines and rules, but work with their patients in a much more flexible way and find the way to create this kind of true healing environment.

 

[Enid]

I do wish the Neurologists would get engaged more (especially infections). My own after extensive investigations on obvious neurological impairments (including brain MRI "high spots") gave up and sent me a letter of apology. If they were more informed of ME pathologies - albeit crossing discipline bounderies - they seem an obvious choice for diagnosis and treatments - antivirals/biotics etc. in the first place. Especially now cancer specialists in Norway found something of an answer to ME - infection.

 

[Esther12]

re the personality stuff: Nater was involved in a really poor study on this, that was then spun to the media. It looks like the same thing is being done here. They failed to mention the prospective Wessely study which found no association between CFS and personality abnormalities, and was a follow up to the Kato study they made a big deal of. They also fail to mention the numerous studies showing that personality problems in CFS are similar to those found in conditions like MS. They've just picked the studies they want to mention to create the narrative that they want to promote... these sorts of things can be so trusted by readers, despite having so little bearing to reality.

There were a few letters about the recent Nater personality study that never got published by the journal. Here's one of them [one other removed on request], and I'm sure others could be found:

Personality studies of the chronically ill may do more harm than good.

In their recent article, Nater et al. [1] compare the answers on two personality questionnaires [2,3] that were provided by Chronic Fatigue Syndrome (CFS) patients, controls with disruptive but less serious ‘insufficient fatigue’ (ISF) and healthy controls. Perhaps unsurprisingly, the two groups suffering from health problems agreed to statements such as “I expect other people to do favors for me even though I do not usually do favors for them” at a greater rate than healthy controls. This led to the diagnosis of personality disorders in 29% of CFS patients, 28% of ISF patients and 7% of well controls.

The impact that living with a chronic illness has upon an individual’s life could alter their assessments of themselves in many different ways. For example, Nater et al. report that a quarter of CFS patients are unemployed or on disability benefit; this might be expected to increase the likelihood of patients agreeing to the statement “I see myself as thrifty, but others see me as being cheap”[2], without being an indication of a greater prevalence of obsessive-compulsive personalities. The use of personality questionnaires to diagnose personality disorders amongst the chronically ill may lead to misleading results if such confounding factors cannot be accounted for. Nater et al. note (p316), “studies comparing CFS to other chronic illnesses, such as multiple sclerosis, did not find differences regarding the prevalence of comorbid personality disorders” and this result would seem to be repeated again here.

Coverage of Nater et al.’s work [4,5,6] followed a news release titled ‘Is Chronic Fatigue Syndrome A Personality Disorder?’[7], and promoted the notion that those suffering from CFS are liable to be suffering from severe personality abnormalities related to non-compliance with treatment suggestions. Given the design of the study and the manner of its presentation, such coverage is not surprising. While Nater et al. focus upon the possibility that it is CFS patients’ personalities which lead to a damaging lack of trust and confidence in health care professionals, it could be argued that these are reasonable responses to the negative stereotyping and stigmatization which patients already face [8,9,10], and which papers that casually emphasise the role maladaptive personality features may play in CFS are likely to worsen. Such problems are especially worrying given the concerns raised about the broad criteria the researchers used to define CFS, which appear to lack both sensitivity and specificity. [11,12]

While Nater et al. note many of the limitations of their work, and chose to include a control group suffering with health problems which was absent from their initial presentation of results [13], great care needs to be taken with the design and presentation of such studies in order to ensure meaningful results are gathered, and that they are not used to legitimise the crude profiling and stereotyping of patients. Otherwise, patients agreeing with the statement “Others will use what I tell them against me” [2] might be displaying prescience, not paranoia.

[1] Nater UM, Jones JF, Lin JM, Maloney E, Reeves WC, Heim C: Personality features and personality disorders in chronic fatigue syndrome: a population-based study. Psychotherapy and Psychosomatics. 2010;79(5):312-8

[2] Hyler SE: Personality Diagnostic Questionnaire-4+ (PDQ-4+). New York, New York State Psychiatric Institute, 1994.

[3] American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 4, text rev (DSM-IV-TR). Washington, American Psychiatric Association, 2000.

[4]‘Is Chronic Fatigue Syndrome A Personality Disorder?’ rehacare: http://www.rehacare.com/cipp/md_reh..._Fatigue_Syndrome_A_Personality_Disorder.html Accessed November 21, 2010

[5] ‘Is Chronic Fatigue Syndrome A Personality Disorder?’ Medical News Today: http://www.medicalnewstoday.com/articles/207006.php Accessed November 21, 2010

[6] ‘Tired and Tested’, Metro. November 9, 2010:27 http://e-edition.metro.co.uk/2010/11/09/ Accessed November 21, 2010

[7]‘Is Chronic Fatigue Syndrome A Personality Disorder?’ Journal of Psychotherapy and Psychosomatics, AlphaGalileo: http://www.alphagalileo.org/ViewItem.aspx?ItemId=89186&CultureCode=en Accessed November 21, 2010

[8] Drachler Mde L, Leite JC, Hooper L, Hong CS, Pheby D, Nacul L, Lacerda E, Campion P, Killett A, McArthur M, Poland F. The expressed needs of people with chronic fatigue syndrome/myalgic encephalomyelitis: a systematic review. BMC Public Health. 2009 Dec 11;9:458.

[9] Patients with chronic fatigue syndrome are not treated properly BMJ. 2004 June 5; 328(7452): BMJ doi: 10.1136/bmj.328.7452.0-e

http://www.bmj.com/content/328/7452/0.6.full. Accessed November 21, 2010

[10] Newton JL, Jones DEJ. ‘Making sense of fatigue’ Occupational Medicine 2010; 60(5): 326-329

[11] Kindlon T. Criteria used to define chronic fatigue syndrome questioned. Psychosom Med. 2010;72:506-7 [12] Jason LA, Evans M, Brown A, Brown M, Porter N, Hunnell J, Anderson V, Lerch A. (2010). Sensitivity and specificity of the CDC empirical chronic fatigue syndrome case definition. Psychology. 2010;1:9-16.

[13] Nater UM, Jones JF, Lin JM, Maloney E, Reeves WC: Abstract 1747: The Role of Personality in Chronic Fatigue Syndrome: Findings from a Population-Based Study. 2009. American Psychosomatic Society 67th Annual Meeting.

 

[Snow Leopard]

As usual they've ignored the sociological aspects of 'dysfunctional illness beliefs'. The problem is that in an absence of good science (due to lack of research), given the stigma and the fact that societal understanding of this illness is extremely poor and this carries right through to many medical practitioners themselves who are also under the veil of ignorance.
If there was good understanding and good treatment available, the rate of 'dysfunctional illness beliefs' would be far lower.
I haven't looked into it, but I bet there was a similar trend in HIV/AIDS for example, as treatment quality and awareness improved, patient beliefs likely also became more realistic.

 

[Enid]

Big problem "beliefs" - (ripe area for the mumbo jumbos waiting in the wings in the absence of science - willful ignorance really). It is unbelievable that in 2012 one can collapse in Accident and Emergency and be told by 4 junior Docs and a psychiatrist that it is all in my mind. Where are we back with the witches. Let's make it perfectly clear theirs is their belief - mine is knowing illness and escaped by the skin of my teeth. Come on Neurologists (certainly mine) knew more but not enough yet. Falling about and passing out is hardly a belief.

Gee whiskers are these "belief" experts idiots. Before I roll over and become very rude with unprintable matter.

 

[biophile]

Can anyone think of a single double-blind trial for CBT or GET? I don't see how such a thing could be done. Or single blind! Which is a bit of a problem when people are relying upon subjective questionnaire scores as outcome measures, particularly when these have been shown not to correlated well with objective measures of disability.

The PACE Trial was unblinded, and AFAIK, there are no "double-blind" RCTs for individual CBT or GET as claimed by Nater et al, but I managed to find one for group CBT (http://www.ncbi.nlm.nih.gov/pubmed/17014748), although the exact quality of blinding is unclear. Also, there is a GET plus fluoxetine study (http://www.ncbi.nlm.nih.gov/pubmed/9828987) where "drug treatment was double blind and patients were blind to assignment to exercise or appointments".

However, CBT is not like a medication, simply removing the label by using a "double-blind" RCT would not rule out all potential cognitive biases introduced by CBT itself. Objective data would remain important despite blinding. It would still be difficult to distinguish altered questionnaire-taking behaviour from genuine reductions in subjective symptoms, especially if the effect sizes are small. Of course, some proponents may argue that there is no important distinction.

Removing the label from GET probably wouldn't make much difference either, although proponents may argue that advocates have tainted the GET label which could result in a nocebo effect. A patient may not know which arm of a trial they are in, but how can they be adequately blinded from exercise they perform themselves?

Another bias to control for is differences in expected outcomes. The PACE Trial failed to do so, the CBT and GET participants were not only given more optimistic initial expectations than the APT and SMC participants, but this was also persistent enough to cause conditioning over time. These expectations are part of the underlying rationale for the respective therapies, but it would have been possible to remove expectations from the manuals. On the other hand, CBT still tells patients that post-exertional symptoms are harmless and due to deconditioning. Also, some people believe that placebo response/effect and improvements in CFS are the same due to the (proverbial) mind-body connection.

When I Googled for this post, I found this study which I've seen before (http://www.racgp.org.au/afp/201111/201111kkaspers.pdf), it was a study of the medications that patients used, and also conducted a literature review of these same medications for comparison:

We excluded trials that were not randomised, lacked blinding or without a control group. The methodology of RCTs minimises the risk of known and unknown biases, including the placebo effect, the influence of the characteristics of the treating doctor, patient preferences and disease severity.

It would be amusing if the next Cochrane reviews of CBT/GET took the same approach. The much promoted "evidence base" of CBT/GET would be reduced to 2 small studies. The PACE Trial would not be one of them, despite being praised as the highest grade of clinical evidence by the Science Media Center's "expert opinion" press release.

FWIW, here is a summary of the unimpressive results of the group CBT study mentioned above:

Summary

The complex nature of this illness, the paucity of current high-quality research and the pragmatic and clinical difficulties of running a trial in this area all contribute to a highly complex analysis and interpretation. We have presented all of our data, with explanations of our reasoning, so that the work is as transparent as possible. The principal outcome measure, the SF-36, did not demonstrate any difference in the number of subjects whose physical function scores returned to within the normal range at the end of the trial. However, there were significant improvements in the CBT group in the measures used for fatigue, mental health and walking speed. The treatment did not, therefore, restore ‘normal’ levels of physical function on the SF-36, but did produce improvements in three other key areas.

The group CBT significantly increased measures of mood and fitness and it significantly decreased symptoms of fatigue. This was comparable to the changes seen in the individual research literature. In Whiting and colleagues’ review, studies were classified as having an overall effect if they showed an effect for more than one clinical outcome.[17]

Group CBT did not significantly improve cognitive function, quality of life (as measured by the physical subscale of the SF-36), employment status or healthcare utility measures. However, the quality of the data obtained for the last item was too poor to be used in any analysis. Such changes have been demonstrated in the literature for individual CBTs.

The outcome measures showed a consistent trend towards improved health status for the research population randomised to CBT. The group did not, however, return a significantly higher proportion of subjects back to the ‘normal’ range of physical function on the SF-36 than the other treatment interventions. It proved too difficult to provide an adequate assessment of the costeffectiveness of the intervention and the reasons for this are discussed below.

http://www.hta.ac.uk/fullmono/mon1037.pdf