New PACE paper, more SMC spin.

[Woolie]

@Bob, not sure if you saw this, but this metanalysis, posted by @chipmunk1 on another thread, also addresses the issue of objective vs. subjective outcome measures:

Cognitive behavioural therapy for major psychiatric disorder: does it really work? A meta-analytical review of well-controlled trials
http://www.academia.edu/307214/Cogn...a-analytical_review_of_well-controlled_tria

This is a metanalysis of studies of the the effectiveness of CBT on major psychiatric disorders (schizophrenia, major depression, bipolar). But uniquely:

* They selected only studies with objective measures (for example, ratings of symptoms made by someone other than the patient, noting whether or not the rater was blind to the intervention).

* They selected only studies with some sort of control condition, that at least had the appearance of a treatment, but wouldn't be expected to be as effective as the CBT (befriending, relaxation sessions, etc.)
​

Guess what, virtually no effects of CBT! A small effect for major depression, but only when the control condition was a placebo pill (not when it was something involving regular human contact).

The point isn't whether or not CBT is effective for this that or the other, The point is that it illustrates the dangers of not applying the usual medical study quality standards to assessments of psychological interventions.

 

[Sean]

Guess what, virtually no effects of CBT! A small effect for major depression, but only when the control condition was a placebo pill (not when it was something involving regular human contact).

I recall reading a general media article a few years back where a psychiatrist argued most patients would be better off just getting a dog then undergoing any form of psychotherapy (drugs or talking).

 

[akrasia]

Notice how, in the narrative, they neatly link the issue of ME/CFS patients' psyches to the alleged death threats, implicitly suggesting that the one or two alleged death threats somehow reflects on the entire ME/CFS population and therefore reflects on a ME/CFS diagnosis. I think that's what's known as a slur on a patient community.

The terrorist/psycho slander is a useful distraction from the primary narrative, the truth that 250,000 people are sick with something that's not understood. Normally, the fact that so many are horribly ill would be seen as a crisis demanding action but if the patients are perpetuating their illness through what they do then the crisis is camouflaged and disappears in the fog of blame and pseudo responsibility. It becomes only logical that nothing needs to be done.

The canard about threats to researchers is just more proof that the original framing of the illness as psychological is correct. Sort of gilding the lily.

I believe what is going on here is basically a rear guard action. One day, and I hope it is soon, the history of m.e. in Britain will look like the stitch up that it is.

 

[Sean]

While any sort of threat is unacceptable, on the other hand, it is incumbent on people in power to make sure those they have power over are not treated in a way that leaves them with little left to lose.

If you got a wild lion loose in the yard you got no right to complain about the consequences if you willingly leave the safety of your house, and repeatedly poke it in the eye with a very short stick.

For people supposedly expert in human psychology and behaviour, they sure can be a bunch of effing numpties sometimes.

 

[Countrygirl]

The Economist now runs with the story...............

http://www.economist.com/news/scien...ld-insights-fear?fsrc=scn/tw_ec/fear_to_tread
The results, which are the subject of the new paper, show one of the main ways CBT and GET trumped APT was that people prescribed them were less fearful than those prescribed APT. This seemed to account for about 60% of the advantage they brought. GET also brought an increase in tolerance to exercise that APT, with its concern not to overstep the boundaries, did not. One psychological mechanism, then, and one physiological one.

That both are involved is hardly surprising. The link between a healthy mind and a healthy body (and vice versa) has been known since classical times. That, though, is by no means the same as saying something is all in the mind.

From the print edition: Science and technology

 

[Bob]

The Economist now runs with the story...............

It's amazing... It's like lifting a rock in the garden and you see all these journalists crawl out, climbing over each other competing to pump out this churnalism garbage... But where are they the rest of the year when we want them to write decent articles about decent stuff that's going on in the world of ME? Festering under the rock I guess. I'm beginning to think that we all need to get together and try to write some articles for publication. It seems like it might be the only way we can compete.

 

[worldbackwards]

You are clearly not placing the right empharsis on the sillarbles.


When it comes to making a fool of myself, I am fearless!

To be fair, my joke about it is a lot worse

 

[Bob]

@Woolie @Sean thanks again for those papers re placebo, etc. And thanks for the helpful summaries, Woolie.

 

[Sean]

To be fair, my joke about it is a lot worse

Pfft. We are amateur punsters. You just wait until the real pro arrives.

Over to you, Graham...



---------------

Cheers, Bob.

I think that placebo paper is one of our core generic arguments. It both places serious upper limits on what can be claimed based on the placebo effect, and highlights a serious methodological limitation in studies that don't have a no treatment, no placebo arm.

Plus the finding (by Wessely himself, no less,) that ME/CFS patients seem to have a reduced placebo effect. (Though I think that needs more study).

As it stands, there is not much wiggle room for them via placebo effect.

 

[Roy S]

Thank you. @chipmunk1

So The Telegraph published a story in the science section by Sarah Knapton, science editor, at 12:01 when the embargo was lifted; then six hours later published a story in the health section by an anonymous "Telegraph Reporter", that has the most objectionable statement that most people and groups are reacting to; including the NHS news:
　
"The Daily Telegraph's suggestion that the study says "people suffering from ME [myalgic encephalopathy] should get out of bed and exercise if they want to alleviate their condition" is particularly unhelpful, and feeds the idea that people with CFS are "lazy": this is not the case."
　
http://www.nhs.uk/news/2015/01January/Pages/Therapy-and-exercise-may-help-some-with-CFS.aspx
　
From the perspective of someone from the rebellious colonies this whole thing is fascinating and more than a bit bizarre. I wonder if the Science Media Centre will again state publicly like they did in 2013 that --
　
"The SMC engineered the coverage"

"the SMC has also helped to set the agenda and frame the narrative of reporting "

http://forums.phoenixrising.me/inde...cle-in-sunday-times.23050/page-10#post-360792
　
　
　
　

 

[chipmunk1]

The results, which are the subject of the new paper, show one of the main ways CBT and GET trumped APT was that people prescribed them were less fearful than those prescribed APT. This seemed to account for about 60% of the advantage they brought.

Depends on how you interpret it.

The advantage is rather small so 60% of something rather small is not much. So fear is only a small fraction of the problem.

 

[Sidereal]

What's 60% of nothing?

 

[chipmunk1]

What's 60% of nothing?

a news story.

 

[biophile]

What's 60% of nothing?

Damn, you beat me to it! I was going to say ...

This seemed to account for about 60% of the advantage they brought.

I haven't spent much time on this yet but it seems that somehow the 20% total mediation effect mentioned at conference presentations magically grew into 60% for (single factor?) mediation in the published paper?

2011: "The [instrumental variable methods]s were weak, with a maximum R^2 change of 0.03. The five strongest IVs were therefore used in the 2SLS in each case. There was modest mediation of CBT and GET effects (approximately 20% of the total effect). The IV-derived estimators were somewhat different in magnitude than the BJK estimators and were less precise. There is scope for modelling a common effect of mediators on outcomes across trial arms."

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3287720

2015: "Fear avoidance beliefs, the strongest mediator, accounted for up to 60% of the overall effect, providing evidence for partial mediation. Many of the mediators accounted for much smaller proportions of the overall effects, suggesting that in some cases the effects of treatment on outcomes might have been mediated through several small effects and that some of the overall treatment effects were mediated through variables that were not measured."

Again, I haven't looked into this paper much yet and some of the statistics are probably beyond me but someone else pointed out that the "planned" analysis type isn't mentioned in the protocol or statistical analysis plan. Could 4 years of post-hoc protocol tinkering and data dredging be at play here? Most of their previous papers greatly deviated from protocol too.

60% of a modest improvement on subjective measures is still small. The nature of the reported improvement is in question (non-blinded, participants encouraged to view symptoms differently and told treatment was very effective).

One psychological mechanism, then, and one physiological one.

What physiological mechanism? Physical fitness wasn't improved. Improvements were subjective.

No statistically significant improvement for almost all of the objective measures. The editorial implies that the small improvement on the walking test was due to pushing harder on testing (reduced "fear") rather than increased fitness. No data published about patients reporting symptom exacerbation after this test. Can someone remind me what 60% of nothing is? ;-)

Let's not forget the icing on the cake: "Consequently, a temporal separation between the mediator and outcomes was missing, making it difficult to ascertain the direction of the causal relation." So in other words, the direction of causation is unknown and could actually be reduced symptoms causing reduced fear. No mention of that huge caveat in the hype!

 

[user9876]

Damn, you beat me to it! I was going to say ...



I haven't spent much time on this yet but it seems that somehow the 20% total mediation effect mentioned at conference presentations magically grew into 60% for (single factor?) mediation in the published paper?

They switched techniques which maybe says something about the quality of the theory behind mediation analysis in the first place. If different techniques get different results is there a good sound theoretical basis behind any of this. Given the papers I've read so far my belief is that there isn't.

 

[biophile]

Did the Guardian et al. learn NOTHING from last month?

http://www.theguardian.com/science/...-health-news-hype-press-releases-universities

Science and health news hype: where does it come from?

Our research shows that most exaggeration in health-related science news is already present in the press releases issued by universities. As alarming as this is, it creates an opportunity to foster more accurate journalism

Wednesday 10 December 2014 18.00 AEST

 

[chipmunk1]

It should be 'fear-avoidance beliefs' from Lethem 1983 and Waddell 1993 (see Wikipedia). Without the hyphen it is incomprehensible, with it not much better. Even the explanation on WIkipedia is a confusion. These people do not actually have coherent ideas it seems. (Surprise?)

http://en.wikipedia.org/wiki/Fear-avoidance_model

The fear-avoidance model is a model that describes how individuals develop chronicmusculoskeletalpain as a result of avoidance of activities based on fear.[1][2] Introduced by Lethem et al. in 1983, this model helped explain how these individuals experience pain despite the absence of pathology.[2][3][4] If an individual experiences acute pain and manages the situation by using avoidant behavior, a lack of pain increase would reinforce this behavior.[5]However, this avoidant behavior may cause the individual to decrease exercise which may in turn lead to increased disability.[5] In 1993, Waddell et al. developed a Fear-Avoidance Beliefs Questionnaire (FABQ) which showed that fear-avoidance beliefs about physical activities are strongly related to work loss.[5][2][6

So it seems to be based on these operant conditioning and reinforcement ideas from pigeon experiments.

 

[biophile]

If only I could get over my fear of breaking my neck I could do a wheelchair double back-flip like this guy?

http://i.crackedcdn.com/phpimages/photoshop/7/3/0/358730_v1.gif

 

[A.B.]

Has there been research to see whether the fear-avoidance model describes reality? Or is it just the usual fantasizing and correlation interpreted however fits their beliefs.

 

[user9876]

http://en.wikipedia.org/wiki/Fear-avoidance_model

So it seems to be based on these operant conditioning and reinforcement ideas from pigeon experiments.

Its a really strange model. Firstly it assumes pain is absent without activity. Secondly it assumes only one motivating factor in a decision. Even if there were such thoughts there would be other thoughts motivating or necessitating activity and these would need to be included in a model. A model should also acknowledge that other effects may support say lying down (for example the paper discussed on another thread http://forums.phoenixrising.me/index.php?threads/increasing-orthostatic-stress-impairs-neurocognitive-functioning-in-cfs-with-pots.12274/ about cognitive abilities for people with ME and OI where cognitive performance gets worse at higher tilts on a tilt table).