Discussion in 'General ME/CFS News' started by Cheshire, Oct 1, 2016.
While the article is meh for us, it has attracted some great comments, including some from Julie Rehmeyer and James Coyne.
Small effect size on subjective self-report questionnaires only, which disappears at 2.5 years...
My brain is a bit fried again, someone should point out the 2.5 year follow-up paper with a reference and show that they don't use the word "null" even though it affirms the null hypothesis.
Comments on this article are handled by the website 'plugin' Disqus, which is used by quite a number of sites, for example I believe the BMJ site uses it. Sign up is easy and I can't recall receiving any spam from signing up, once you've done that you can comment and vote on comments on this article and on others elsewhere in the future.
There is a comment pointing this out already, although with no reference.
I haven't got the energy to sort out a comment myself, but I wonder whether someone could point out that although on average the GET group walked a bit further than the SMC group at I year, about 10 more people in the GET group didn't do the second walk. (I make it 49 and 39). This should be taken into account in some way since it seems likely that at least some of these failed to do the walk because they'd got significantly worse. Someone's done a graph showing those who didn't walk had lower SF36 scores. This makes the comparison of mean walking distance dubious at best. See thread on Pace data.
And a really good one from @Jonathan Edwards
I am still wondering why some journalists trying to be optimistic about a flawed trial. If this was about a cancer drug everyone would be angry and upset. It looks like they can write and say anything about CFS patiënts even if it is scientifically bullshit. The conclusion of any rational statistician would be: there is a very, very, very small insignificant positive effect for a very, very, very small group of patiënts due to placebo effect called 'pleasing the doctor'. Everybody with a scientific view who argues this conclusion is or grazy or delusional.
He does not GET it.
Loading Tweet... https://twitter.com/statuses/781993135991816192
This line has stuck in my head and would make a good reply headline to Whites article in the Guardian:
'PACE study definition of recovery of ME/CFS is worse than an average 80-year-old'
Jonathan Edwards comments were good and this one:
Doc Mary • 11 hours ago
I was always taught that when you are at a decision point in a study, you should "argue against your own thesis." When it comes to Dr. White, every choice he has made has biased the results - and his reporting of the results - towards the outcome in which he has long been invested. When they decided to change the outcome measures, doing so FAVORED their position.
Borrowing from Bruce Levin's comment (as posted here by Julie Rehmeyer), when they changed the outcome measures, they could have reported BOTH results - the result with the original measures AND the result with the measures after the switch. By choosing only to report the latter - AND not telling anyone what the result would have been had they gone with the former, once again, they made a choice which biased the results in a way that FAVORED their position.
When they reported the results to the public, they again used an interpretation that heavily FAVORED their point of view. 60% of patients recovered! (Telegraph) WOW! This must be great. No. It's wrong - now ridiculously so.
Indeed, given that they spent the better part of 20 years (when they published the study), the CBT/GET proponents have made this the ONLY treatment offered patients in the UK through NHS. CDC promoted it in the US.
In his position as Chief Medical Officer of Scottish Provident, Peter White even advised the insurance company not to give anyone disability unless they had undergone a ten-week program of CBT/GET. If they could not complete it, they could not get disability.
It is that latter part that has made this whole story so tragic. If the question was whether CBT/GET was effective or not, to some extent the data manipulation that occurred (and yes, Mr. Neuroskeptic, changing the goalposts in the middle of the game is considered data manipulation) would be interesting, but not tragic.
Unfortunately, CBT/GET - specifically GET (Graded Exercise Therapy) - has done great harm to patients with ME in the UK. A poll by the ME Association found that 74% of patients reported that a course of Graded Exercise had made them significantly worse.
And there are logical reasons why graded exercise - forcing a patient to under an exercise regimen where they increased their daily exercise by a fixed amount, ignoring however it made their bodies feel - is dangerous for patients with ME.
Researchers in multiple universities in the US, beginning with Mark Van Ness, Christopher Snell, and Staci Stevens, have found that (higher functioning) patients with ME who undergo CPET (or VO2 max testing), score similarly to a sample of "couch potatos" - deconditioned but healthy individuals. But on a SECOND day of testing, the deconditioned controls scored the same (actually they improved slightly) - while the ME patients deteriorated by as much as a half.
A patient whose CPET scores drop continuously as they exercise should not be in a program of graded exercise. Period. It could be predicted that many would end up sicker, even permanently sicker - which is exactly what patients have been saying for 25 years.
Another recent study at UC San Diego and studies by Maureen Hanson at Cornell University suggest that there are serious metabolic deficiencies in patients with ME/CFS (Canadian 2003). These abnormalities, again, suggest that blindly requiring patients to exercise could cause great damage.
But these are biomedical studies, and you will find no footnotes of biomedical studies in the writings of the biopsychosocial school of medicine to which Dr. White and his comrades belong. Their work exists in a parallel universe, where they cite each other's work and nothing else.
It is also worth noting at this point that Dr. White and his colleagues also tilt the study in their direction by using the Oxford definition for "CFS/ME" - a definition which requires six months of debilitating fatigue with no physiological explanation. In insurance language lingo, that means the patients have "MUS" - medically unexplained symptoms." That requirement, that there be no medically explainable physical symptoms, goes against every other definition for CFS or ME used in the world today. The two most common in current research are Fukuda (1994), used by CDC:
and the Canadian Consensus Criteria (2003), created by committee when Canada adopted ICD-10-CA, which placed CFS with ME in the tabular version at G93.3. [ME has always been coded in the ICD's in the chapter on neurology and remains at G93.3 in ICD-10.]:
[There are many other definitions currently in use, all more complicated than Oxford.]
The main differences between Oxford and EVERY OTHER DEFINITION in use in the world, are (1) Oxford requires only debilitating fatigue, and (2) Oxford does not exclude mood disorders. As a result, any study conducted using Oxford is going to pull in patients with primary mood disorders, not primary ME or CFS. It then becomes confusing how to interpret the results.
A study commissioned by the US NIH from its P2P program concluded that the Oxford definition should be retired. The Agency for Healthcare Quality and Research, also within NIH, went a step further this summer and added the recommendation that studies CONDUCTED WITH THE OXFORD DEFINITION be discarded. Since the only studies with positive results for CBT/GET were conducted using Oxford, AHQR concluded that they could no longer recommend CBT/GET as a possible treatment for ME/CFS.
in other words, this study is outdated, period - without the data manipulation.
Those who know only the research conducted by this small group of psychiatrists might find the following a useful bibliography of research into biomarkers for ME/CFS:
I would suggest that as a way to BEGIN to study the disease ME, given that the scientific literature long ago bypassed the "biopsychosocial" psychiatrists behind this deeply flawed study.
One more thing: Given the evidence that graded exercise can be very dangerous for patients with ME or a diagnosis of CFS, the planned study for children and teenagers, called MAGENTA, should be halted NOW, rather than risk permanent harm to the most vulnerable among us."
Shame this couldn't go in the Guardian.
We need to be aware of what a big thing it is for UK researchers in this area to recognise that PACE is 'bad science'. It's disappointing that Neuroskeptic publicly does otherwise, and gives such poor justifications for it, but a lot of those in research are going to have a strong desire to avoid recognising the legitimacy of our criticism.
Thanks a lot to @Jonathan Edwards for his comment and all his work on PACE.
Yes - I don't know much about neuroskeptic - is he afraid to admit the truth or what - does he have past form on this topic that prevents him from opening his eyes or what? Conflicts of interest ? Just speculating.
NeuroSkeptic - not so skeptic after all?
It wouldn't amaze if it was just emotionally unsettling to come out against authority in the way classing PACE as bad science would require. Wessely, the MRC... a lot of big people are fully behind this.
I genuinely suspect he misses the point. It is a bit like saying Charles Ives's Fourth Symphony should not be called bad because there are enough B flats in it. Science occurs in the messy context of real people with strange social conventions. A Vulcan analysis may come unstuck.
Henrik Vogt added some comments, well, hum, à la Henrik Vogt.
He has a unique charm.
I would like to thank you for all your replies and your courage to stand up for all CFS/ME patiënts! THANKS!
You can also try a Google Site Search
Separate names with a comma.