New PACE paper, more SMC spin.

[Sean]

It is the total lack of understanding of carer/patient relationship that I find so mind-boggling - and they are supposed to be experts in human nature!

It is truly bizarre and, frankly, terrifying that the profession who should know all about the dangers of viewing and treating people in this way, are the very ones viewing and treating people in this way.

 

[Roy S]

The Economist
Chronic Fatigue Syndrome
Fear to tread
A controversial trial on a mysterious disease continues to yield insights
　
This is in the text referring to the brouhaha after the original PACE paper was published:

"Many researchers studying psychological or psychiatric causes of or treatments for CFS reported being harassed in person and online. Some received death threats."
　
"Fear to tread"? I think they're walking all over us. Again. Well played, SMC.
　
http://www.economist.com/news/scien...ld-insights-fear?fsrc=scn/tw_ec/fear_to_tread
　
　
　
　

 

[Sidereal]

Economist:

The findings sparked a furious and public debate between the researchers, the editors of the Lancet, advocacy groups and some patients, who called the study’s methods and conclusions into question. At issue was the quantity of resources spent looking at effects in patients’ psyches rather than hunting for more tangible causes and effects in their bodies. Many researchers studying psychological or psychiatric causes of or treatments for CFS reported being harassed in person and online. Some received death threats.

 

[Roy S]

The Telegraph article has been altered AGAIN. It now says:
"Fear of exercise exacerbates ME and sufferers need to try and get up out of bed if they want to get better, a major study by King's College has found."

The Telegraph article has been altered to delete the first statement under the photo reiterating some of the disgusting 2011 propaganda campaign organized by the Science Media Centre blaming activists for harming research. Did anybody save that part?

This bit of [hogwash] is still the lead:

"People suffering from ME should get out of bed and exercise if they want to alleviate their condition, a major study by King's College has found."


My commiserations again go to those of you who have to live in the middle of this.

 

[chipmunk1]

"People suffering from ME should get out of bed(writing death threats) and exercise if they want to alleviate their condition

 

[Esther12]

The Telegraph article has been altered AGAIN. It now says:
"Fear of exercise exacerbates ME and sufferers need to try and get up out of bed if they want to get better, a major study by King's College has found."

lol - are they just changing things at random?

 

[chipmunk1]

the telegraph published 2 different articles on the same day one under science and one under health.

maybe that explains some of the confusion:

http://www.telegraph.co.uk/news/sci...chronic-fatigue-syndrome-say-researchers.html

http://www.telegraph.co.uk/news/health/news/11343681/Chronic-fatigue-syndrome-the-symptoms.html

the latter still mentions:

The minority of militant sufferers are doing the majority a tremendous disservice by scaring off doctors and scientists from working in the area

 

[Bob]

For reference:

People suffering from ME should get out of bed and exercise if they want to alleviate their condition, a major study by King's College has found.

Fear of exercise exacerbates ME and sufferers need to try and get up out of bed if they want to get better, a major study by King's College has found.

 

[worldbackwards]

lol - are they just changing things at random?

Ooooh, can I try the next one? Projected for the morning edition:

Fear of King's College exacerbates exercise and ME sufferers need to try bedwetting if they want a major study by better people.

 

[akrasia]

@Roy S wrote:

"Fear to tread"? I think they're walking all over us.

Yes, but The Economist in its bid to be clever omitted the first part: Fools rush in...
　
　

 

[Sean]

"Many researchers studying psychological or psychiatric causes of or treatments for CFS reported being harassed in person and online."

In triple time, and everybody join in...

Pink elephants in the room
dropping el primo bullshit,

Flying pigs and trolls
and no need to prove it

Bright shiny knighthoods,
recovery for all

These are a few of my favourite frauds.

 

[worldbackwards]

Economist:

The findings sparked a furious and public debate between the researchers, the editors of the Lancet, advocacy groups and some patients, who called the study’s methods and conclusions into question. At issue was the quantity of resources spent looking at effects in patients’ psyches rather than hunting for more tangible causes and effects in their bodies. Many researchers studying psychological or psychiatric causes of or treatments for CFS reported being harassed in person and online. Some received death threats.

It's, ahem, interesting, that no-one wants to engage with the accusations of poor science. One might almost think that they were looking for something that they had a ready made answer for.

 

[Woolie]

Does anyone know of a research paper that demonstrates the effect size from a placebo intervention?

This is one of the most influential papers on the size of the placebo effect, its a metanalysis. https://math.dartmouth.edu/~doyle/docs/40/PlaceboArticle.pdf

Although they don't call it that, the measure they call pooled standardised mean difference is effectively an effect size measure (similar to Cohen's d). The study combines a number of different conditions and treatments, most of which have a psychological component (obesity, depression, smoking cessation, anxiety, etc), so if we'd expect to see placebo effects anywhere it would be here.

Here are the overall effect sizes, depending on whether outcomes was measured on a subjective scale (e.g., self report) or an objective scale:

Subjective -0.36 (confidence interval -0.47 to -0.25)
Objective -0.12 (confidence interval -0.27 to 0.03) This is a null effect.

So no effect on objective measures, even in these illnesses that might b expected to be highly responsive.

From the discussion:
"It is difficult to distinguish between reporting bias and a true effect of placebo on subjective outcomes, since a patient may tend to try to please the investigator and report improvement when none has occurred. The fact that placebos had no significant effect on objective continuous outcomes suggests that reporting bias may have been a factor in the trials with subjective outcomes."

 

[Bob]

Economist:

At issue was the quantity of resources spent looking at effects in patients’ psyches rather than hunting for more tangible causes and effects in their bodies. Many researchers studying psychological or psychiatric causes of or treatments for CFS reported being harassed in person and online. Some received death threats.

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.

 

[Sean]

This is one of the most influential papers on the size of the placebo effect, its a metanalysis. https://math.dartmouth.edu/~doyle/docs/40/PlaceboArticle.pdf

A more recent updated version:

http://www.ncbi.nlm.nih.gov/pubmed/20091554

 

[Bob]

This is one of the most influential papers on the size of the placebo effect, its a metanalysis. https://math.dartmouth.edu/~doyle/docs/40/PlaceboArticle.pdf

Although they don't call it that, the measure they call pooled standardised mean difference is effectively an effect size measure (similar to Cohen's d). The study combines a number of different conditions and treatments, most of which have a psychological component (obesity, depression, smoking cessation, anxiety, etc), so if we'd expect to see placebo effects anywhere it would be here.

Here are the overall effect sizes, depending on whether outcomes was measured on a subjective scale (e.g., self report) or an objective scale:

Subjective -0.36 (confidence interval -0.47 to -0.25)
Objective -0.12 (confidence interval -0.27 to 0.03) This is a null effect.

So no effect on objective measures, even in these illnesses that might b expected to be highly responsive.

From the discussion:
"It is difficult to distinguish between reporting bias and a true effect of placebo on subjective outcomes, since a patient may tend to try to please the investigator and report improvement when none has occurred. The fact that placebos had no significant effect on objective continuous outcomes suggests that reporting bias may have been a factor in the trials with subjective outcomes."

Wow, that looks really helpful. Thanks for that Woolie. I'll save it for some bed-time reading.

Very interesting that subjective outcomes change, but objective outcomes don't change - just like the PACE trial!

 

[Woolie]

A more recent updated version:

http://www.ncbi.nlm.nih.gov/pubmed/20091554

Nice addition, @Sean.

For those not up to reading the full pub:

Subjective -0.26 (confidence interval -0.32 to -0.19)
Objective -0.13 (confidence interval -0.24 to -0.02)

From the article: "This update confirms and modifies the findings of the previous versions of our review. Our approach can be seen as testing the hypothesis that placebo treatments have large effects across many clinical conditions and outcomes, and our results clearly indicate that this hypothesis is wrong."

 

[worldbackwards]

In triple time, and everybody join in...

Pink elephants in the room
dropping el primo bullshit,

Flying pigs and trolls
and no need to prove it

Bright shiny knighthoods,
recovery for all

These are a few of my favourite frauds.

You'll have to work harder, Sean, it doesn't scan properly! One might think you'd failed to make sufficient effort because of fear avoidance!

 

[Woolie]

Would it be too bold to suggest that any intervention study that is unblinded (like PACE) should be required to demonstrate an effect size reliably in excess of those found routinely with placebo interventions? We might amusingly call it the "snake oil test"!

There has been a move in experimental psychology lately towards improving the quality and replicability of studies by improving publishing standards. One suggestion is preregistration: before researchers collect data, they publish all outcome measures they will be using; they must then reported the results of each of those outcomes and no others.

Those interested more generally in the problems of behaviorual research and possible solutions, see:

False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant
http://pss.sagepub.com/content/22/11/1359.short?rss=1&ssource=mfr

 

[Sean]

You'll have to work harder, Sean, it doesn't scan properly!

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

One might think you'd failed to make sufficient effort because of fear avoidance!

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