• 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.

PACE trial commentary in JHP from Dr Charles Shepherd, ME Association

charles shepherd

Senior Member
Messages
2,239
PACE trial commentary in JHP from Dr Charles Shepherd, ME Association

The MEA commentary in Journal of Health Psychology on the PACE trial methodology and results has now been published

Open access to full paper here

http://www.meassociation.org.uk/201...d-journal-of-health-psychology-10-april-2017/

PACE trial claims for recovery in myalgic encephalomyelitis/chronic fatigue syndrome – true or false? It’s time for an independent review of the methodology and results

ABSTRACT
The PACE trial set out to discover whether cognitive behaviour therapy and graded exercise therapy are safe and effective forms of treatment for myalgic encephalomyelitis/chronic fatigue syndrome.

It concluded that these interventions could even result in recovery. However, patient evidence has repeatedly found that cognitive behaviour therapy is ineffective and graded exercise therapy can make the condition worse.

The PACE trial methodology has been heavily criticised by clinicians, academics and patients. A re-analysis of the data has cast serious doubts on the recovery rates being claimed.

The trust of patients has been lost.

The medical profession must start listening to people with myalgic encephalomyelitis/chronic fatigue syndrome if trust is going to be restored.


Dr Charles Shepherd
Hon Medical Adviser, MEA
 

Snowdrop

Rebel without a biscuit
Messages
2,933
I confess to being very untrusting of the people who have anything to do with ME outside the community of non-BPS advocates and ME sufferers themselves. So this is why I am suggesting that if an 'independent' inquiry/re-analysis is to take place then I think a discussion ought to be had as to what 'independent' means.

As we know, PACE was never about science but about politics so any 'independent' reanalysis needs defining IMO as to what that looks like.
 

Snowdrop

Rebel without a biscuit
Messages
2,933
@A.B.

Ah, thank-you. I didn't know that.
And I'm glad to hear it. Hopefully that is how it will proceed.

ETA: Although I may be putting my foot in it as I know little of how these things work but my perfect scenario would be a group of known respected statisticians from various countries all having a go. But it may be that stats alone is not sufficient and someone with clinical experience as well needs to be on board.

Regardless of my musings it's good to know that PACE is being exposed for what it is by all those involved.
 

Jonathan Edwards

"Gibberish"
Messages
5,256
I think all some statisticians might do is check the calculations. It needs people with high calibre track record of running clinical trials who understands good scientific method and the dangers of bias etc. This could include statisticians as part of a group.

Agreed, this is not primarily a statistical problem. It is, ironically, a psychological problem. To design a trial well you need to understand the psychology of people who do trials. Psychiatrists seem to be remarkably poor at that. PACE had an official statistician involved in design who seems to have missed the point completely.
 
Last edited:

Snowdrop

Rebel without a biscuit
Messages
2,933
I think all some statisticians might do is check the calculations. It needs people with high calibre track record of running clinical trials who understands good scientific method and the dangers of bias etc. This could include statisticians as part of a group.

Ah yes, I thought something like that might be the case.
 

Snowdrop

Rebel without a biscuit
Messages
2,933
Agreed, this is not primarily a statistical problem. It is, ironically, a psychological problem. To design a trial well you need to understand the psychology of people who do trials. Psychiatrists seem to be remarkably poor at that. PACE had an official statistician involved in design who seems to have missed the point completely.

Oh the irony of that. Anyway, thankfully I'm in good hands with others knowing what is best to do here.
Have at it.
 
Messages
2,391
Location
UK
Agreed, this is not primarily a statistical problem. It is, ironically, a psychological problem. To design a trial well you need to understand the psychology of people who do trials. Psychiatrists seem to be remarkably poor at that. PACE had an official statistician involved in design who seems to have missed the point completely.
I suspect there is a fundamental dilemma with psychiatrists trying to self-analyse ...
 

user9876

Senior Member
Messages
4,556
I have trouble understanding why a competent statistician would allow such obvious flaws in the handling of the basic stats to be let through.

I think many of the statistical issues are ingrained in parts of the medical community. The idea that you can just ask a few questions and then add up the scores and treat it as a scale is so wrong but so pervasive that it seems normal to them. The consequence is that they don't think about the underlying measurements and what they mean just treat them as if they were linear scales.
 

Snow Leopard

Hibernating
Messages
5,902
Location
South Australia
I think many of the statistical issues are ingrained in parts of the medical community. The idea that you can just ask a few questions and then add up the scores and treat it as a scale is so wrong but so pervasive that it seems normal to them. The consequence is that they don't think about the underlying measurements and what they mean just treat them as if they were linear scales.

Yes... Perhaps there is a difference in perspective between those of us who did 100+ experiments during our science undergrad and those who clearly did not... What you think you are measuring isn't always what you are actually measuring. A questionnaire that measures quality of life, or pain or whatever is not quality of life or pain itself.
 

user9876

Senior Member
Messages
4,556
Yes... Perhaps there is a difference in perspective between those of us who did 100+ experiments during our science undergrad and those who clearly did not... What you think you are measuring isn't always what you are actually measuring. A questionnaire that measures quality of life, or pain or whatever is not quality of life or pain itself.

Yes and I think we perhaps have a wider experience than what seems like a quite insular world.

One of the things that economists talk about is proxy measures that is an indirect measure that approximates an unmeasurable so quantity. So something like GDP becomes a measure of standard of living. In this way I've always thought of the questionnaire scores as a proxy measure for the thing they are trying to ask about. Then you need to ask is it a good proxy and what are the characteristics. With the CFQ where they have two marking schemes then it seems strange to me as they don't preserve order. But if you forget the assumption that the questionnaires are just a proxy to a value that cannot be directly measured then you don't realize many of the issues. I suspect much of the problem is doing things by menu rather than thinking out the overall system.

I get the feeling that the statisticians in academia are more interested in their research into things like randomization algorithms, imputting techniques etc and hence don't really give the trial too much thought.