UK:ME Association funds research for a new clinical assessment toolkit in NHS ME/CFS specialist services, 2023

Trish Davis
Perhaps it will help you understand where I'm coming from with PEM if you read the S4ME fact sheet on PEM. https://www.s4me.info/threads/science-for-me-fact-sheets.43310/post-606969

Trish Davis


Sarah Tyson

Trish Davis The results of the PASS gives fascinating and detailed description of PEM in ME/CSF and I look forward to sharing and discussing the details in due course.



Sarah Tyson
Trish Davis If you don’t mind me saying, I think you have not grasped (or maybe are unwilling/unable to accept) that the point of the MEAQ is to ask about what people do, rather than the full detail of how they do it. There will be a multitude of different ways in which people perform tasks, make adaptation or other compromises to get through life. And that is fine. The issue is whether they do it, however they do it. Whether someone else may do it differently, or whether the person can do it in a certain way is irrelevant. We are not attempting to standardise the activities, we are asking if people do the activity, however they do it.

You ask about the scoring system and analysis approach. It is pretty hard to summarise as it is a large section of my life’s work! However, you really want to get your head around these issues I recommend the seminal texts by Anne Bowling. Luckily some of it appears to be available for free now-a-days. This is a good intro (Chapter 2) Measuring Health - Ann Bowling - Google Books
The analysis which focusses on a really thorough examination of the construct validity using Rasch analysis. The Wikipedia page is pretty good introduction. Rasch model - Wikipedia


Trish Davis
Sarah Tyson Thank you for the information about how you analyse the data. I'll take a look.
You say: "If you don’t mind me saying, I think you have not grasped (or maybe are unwilling/unable to accept) that the point of the MEAQ is to ask about what people do, rather than the full detail of how they do it."

My response to that is that I'm sorry to say I find this response patronising and missing my point. As I said earlier, I find FB particuarly difficult to use for in depth discussion, but I've tried to convey the central factors of cumulative effect and how often one can do an activity to gauging key factors of severity and need for care, which are presumably an important part of care planning. However good the mathemeatical/analytical instruments used, any summary statistic is too easily skewed by the sort of data this will provide. I think I'll have to leave it there. I have no energy left for arguing against someone so convinced their knowledge is superior to anything I might say.
 
On the topic of using the word 'avoid'

Sarah Tyson
as I said in a reply to another post, this has nothing to do with the DWP or benefits system. Period. The results of the MEAQ would not appear in anyone's records unless the individual chose to share it.

Trish Davis
Sarah Tyson any information in our clinical records, including our responses to PROMS and care plans based on them, are likely to form the basis of clinicians reports to DWP used as part of our evidence in benefit applications. I can easily see a clinician copying the wording in both care plan and benefit report, and listing activities the pwME 'avoids'. And you say we need to move past the old PACE approach and reclaim language such as 'avoid'. That sounds all very fine until we are faced with the all too common judgements made by a large proportion of clinicians we encounter who do still believe we avoid activity out of fear or laziness. Why feed their prejudices?
 
They could usefully steal a leaf from the PIP book here, and use "cannot do" in the sense of meaning "cannot do without significant adverse consequences".

I think it means the same as what is intended (but left wide open to misconstruction) with "avoid".

Using "cannot" makes it clear that it isn't a choice. When removed from context, the numerous meanings of "avoid" include "can never be bothered to do that, it's so boring".
 
Kitty said:
They could usefully steal a leaf from the PIP book here, and use "cannot do" in the sense of meaning "cannot do without significant adverse consequences".

I think it means the same as what is intended (but left wide open to misconstruction) with "avoid".

Using "cannot" makes it clear that it isn't a choice. When removed from context, the numerous meanings of "avoid" include "can never be bothered to do that, it's so boring".
Click to expand...
I remember the mantra with WCA..
Can you do XYZ?
Yes, I may be not Repeatedly, Reliably Safely....
 
Kitty said:
They could usefully steal a leaf from the PIP book here, and use "cannot do" in the sense of meaning "cannot do without significant adverse consequences".

I think it means the same as what is intended (but left wide open to misconstruction) with "avoid".

Using "cannot" makes it clear that it isn't a choice. When removed from context, the numerous meanings of "avoid" include "can never be bothered to do that, it's so boring".
Click to expand...
If the goal was to actually do this well then sure, as we can point out there are a variety of ways to fix this to avoid the biased language and to address the diseases unique challenges in activities. Having the answers include the impact on other activities for example would drastically improve it and remove the need for explanations, rather than trying to contort with adapation to mean something it doesn't. That doesn't appear to be the goal here, it appears to be to introduce and use this biased language and poor measures intentionally.
 
Sarah Tyson said:
The results of the PASS gives fascinating and detailed description of PEM in ME/CSF and I look forward to sharing and discussing the details in due course.
Click to expand...

I can hardly wait...

Given our detalied critique of the first version, if whatever version they analysed was as badly designed, I'm afraid it will be garbage-in-garbage-out. Not the fault of the efforts of all the pwME who used their precious energy to try to decipher the questions and did their best to fill in the lengthy questionnaire, but of such a badly designed questionnaire it can't possibly achieve valid or useful results.

Prove me wrong, Sarah Tyson...
 
bobbler said:
It will also support goal setting,
Click to expand...

Does any other condition get asked to set goals for their condition?

This is such a rehab mindset; if all you have is a hammer, everything looks like a nail

My goal is simple; regain full health

Doesn't need 3 questionnaires of 90 questions each.

I imagine most other patients with most other conditions have the same goal, but don't get asked epic quantities of pointless questions
 
Amw66 said:
I am not sure that I agree all activities lead to symptoms
Click to expand...

I am not sure that you should be asking pwME questions about it in that case.

Amw66 said:
I assure you it is completely tailored for people with ME
Click to expand...

This reminded me of the Victoria Wood sketch about the woman carrying out a survey on Hellman's mayonnaise.." of course it's totally bona fide - totally fona bido"

I'm always extremely wary of people who say "trust me"

If it is completely tailored for people with me, why are so many people with me pointing out the ways it doesn't work for them and their symptoms..?
 
As a foggy-brained and not at all scientific person, I think the statement about PEM not being cumulative must be the the biggest glaring clanger she has stated (in public, on Facebook)

I’m not going to attempt to argue the minutia, I think this one thing should be taken forward and highlighted.

Does CS agree PeM isn’t cumulative activity?
 
The information from a PROM is used, along with the rest of the assessment to identify the patients’ main symptoms, impairments, activity limitations, and in conjunction with the patient, decide their priorities and goals and devise a treatment plan. That is the point and process of clinical assessment. The effect of the treatment plan needs to be evaluated and monitored – to see if it working and thus whether to continue to whether to change tack.
Click to expand...

Cargo cult stuff. We'll act as if there are treatments, and if we keep acting it out for long enough, maybe the effects of a treatment will happen.
 
The effect of the treatment plan needs to be evaluated and monitored – to see if it working and thus whether to continue to whether to change tack.

Yet again the deep lack of understanding shows through. How does anybody know if the treatment had an effect or whether the patient might even have done better without it. There is no control manoeuvre. So you cannot see if it is working.

You may change tack but the justification for that has to be some sort of measure falling below expectations from controlled trial data that show expected effect magnitudes. (You don't know whether it didn't work but you have some probabilistic indications.) We are nowhere in that ballpark.
 
MrMagoo said:
Bit of a political nightmare isn’t it, who wants to publicly call out the ME A and NHS?
Click to expand...

It is, and that's partly what makes me so hacked off about the direction the MEA has taken and the frankly amateurish work it's now funding.

It puts anyone talking sense into a difficult position. I guess the best they can do is hope the whole farrago is overtaken by events, then fades rapidly from memory.
 
Trish said:
Trish Davis
Perhaps it will help you understand where I'm coming from with PEM if you read the S4ME fact sheet on PEM. https://www.s4me.info/threads/science-for-me-fact-sheets.43310/post-606969

Trish Davis


Sarah Tyson

Trish Davis The results of the PASS gives fascinating and detailed description of PEM in ME/CSF and I look forward to sharing and discussing the details in due course.



Sarah Tyson
Trish Davis If you don’t mind me saying, I think you have not grasped (or maybe are unwilling/unable to accept) that the point of the MEAQ is to ask about what people do, rather than the full detail of how they do it. There will be a multitude of different ways in which people perform tasks, make adaptation or other compromises to get through life. And that is fine. The issue is whether they do it, however they do it. Whether someone else may do it differently, or whether the person can do it in a certain way is irrelevant. We are not attempting to standardise the activities, we are asking if people do the activity, however they do it.

You ask about the scoring system and analysis approach. It is pretty hard to summarise as it is a large section of my life’s work! However, you really want to get your head around these issues I recommend the seminal texts by Anne Bowling. Luckily some of it appears to be available for free now-a-days. This is a good intro (Chapter 2) Measuring Health - Ann Bowling - Google Books
The analysis which focusses on a really thorough examination of the construct validity using Rasch analysis. The Wikipedia page is pretty good introduction. Rasch model - Wikipedia


Trish Davis
Sarah Tyson Thank you for the information about how you analyse the data. I'll take a look.
You say: "If you don’t mind me saying, I think you have not grasped (or maybe are unwilling/unable to accept) that the point of the MEAQ is to ask about what people do, rather than the full detail of how they do it."

My response to that is that I'm sorry to say I find this response patronising and missing my point. As I said earlier, I find FB particuarly difficult to use for in depth discussion, but I've tried to convey the central factors of cumulative effect and how often one can do an activity to gauging key factors of severity and need for care, which are presumably an important part of care planning. However good the mathemeatical/analytical instruments used, any summary statistic is too easily skewed by the sort of data this will provide. I think I'll have to leave it there. I have no energy left for arguing against someone so convinced their knowledge is superior to anything I might say.
Click to expand...
As there is currently another paper including Chalder that has just come up claiming to use Rasch analysis to validate something I am starting to take a poke around and look more closely at this.

Sarah keeps emphasising that the Rasch analysis is the magic somehow making this robust. I've been refreshing myself on this and the critique of its use in healthcare (should only be used where methodology is up to the scratch needed for such an analysis)

For example in Sarah's first para she emphasises the following:
the point of the MEAQ is to ask about what people do, rather than the full detail of how they do it. There will be a multitude of different ways in which people perform tasks, make adaptation or other compromises to get through life. And that is fine. The issue is whether they do it, however they do it. Whether someone else may do it differently, or whether the person can do it in a certain way is irrelevant.
Click to expand...

For laziness' sake I've used AI to give the following description of Rasch more simply (my bold in first para):
Rasch analysis works by converting ordinal (ranked) data from questionnaires or tests into a linear, equal-interval scale called a logit scale. This requires data that is on a scale, where each item represents a different level of the trait being measured. The analysis assesses if the items and persons can be placed on a common scale, with item difficulties and person abilities on a single continuum, making the measurement instrument unidimensional.

How it works
  1. 1. Model-Based Approach:
    The Rasch model is a type of Item Response Theory (IRT) model that evaluates the quality of outcome measures.

  2. 2. Common Scale:
    It creates a common, linear scale (in logits) where both person abilities and item difficulties can be located.

  3. 3. Item Difficulty and Person Ability:
    Items are calibrated to represent their difficulty levels, and persons are scored to represent their ability or trait level.

  4. 4. Invariance:
    The key principle is that these measures are invariant, meaning a person's ability estimate is independent of the specific items used to assess it, and an item's difficulty estimate is independent of the persons who respond to it.

  5. 5. Data Misfit:
    The analysis also checks for "misfit," where the observed responses don't align with the predicted values from the Rasch model, indicating potential issues with the scale's quality or the items themselves.
Click to expand...

I found the example in the following paper simple enough to be useful as an analogy: https://pmc.ncbi.nlm.nih.gov/articles/PMC5132390/#:~:text=Specifically, Rasch analysis allows researchers to use,the unequal difficulties across all test items.

which basically is using the idea of a professor putting together a test paper for students. ANd as part of that having a set of 30 questions, 10 of each at the easier, medium, harder knowledge-level on a very specific subject.

The idea is that one is checking when testing said test by analysis that indeed the results from the test group of students confirm that those 10 thought to be 'easy' are indeed easy and that those deemed hard aren't being answered correctly mostly by those students who got overall low scores, but those who also were capable of doing well on those deemed easy and hard.

This test analogy I would guess works because it fits into the yes/no ie right/wrong answer territory, rather than the 'ness' scale. I can't fathom what the response categories including adjust or avoid are in this PROM survey.


I think the issue with using 'adjust' without a 'ness' aspect to this isn't just that it isn't ordinal but that some activities will be more open to adjustment than others. And the priority of doing something will be individual to someone's circumstance because if you have to take the children to school you are in a different boat to those who don't and might be able to work from home. Immediately people's response to one of the questions is impacted neither by disability nor choice (and are not 'avoiding') if they do not have a reason they are required to take such a trip. So I'm confused how this is interpreted in the assumptions

How for example are the responses that aren't included as an option (such as 'I don't have family', or 'I don't ever have a reason to get on a bus') accounted for in the following part of the survey:

SECTION 5: FAMILY LIFE​



Over the last month, have you done this activity on an average day (rather than a good/best day)? (Choose one answer)


41. Take physical or emotional care of family members (e.g. young children or elders)


42. Have visitors (e.g. friends or family) in your home


43. Form and maintain relationships including those you live with


44. Have intimate relations with your spouse/ partner


45. Is there anything else you would like to tell us about family life?
Click to expand...

and how can anyone guessing answers to these be being consistent - given these would surely have to be how long is a piece of string answers which relate to their hindsight 'points of reference' (what I used to be able to do) or 'imagined situation if they weren't disabled points of reference' or indeed people start incorporating the expectations of those 'others' and whether they are satisfied.

WHat exactly is 'with adjustment' or 'avoid' vs 'yes' on 'form and maintain relationships including those you live with'? for example.

And isn't taking physical or emotional care of family members going to involve very different things depending on the age and situation of for example children or elderly. One person who might be a sole carer for a spouse with dementia might tick 'with adjustment' for taking an hour off when a carer steps in, whilst be undertaking more activity than someone who has a spouse or relative in a care facility that they are visiting once a day for example. Which is different again to having 'taken emotional care' by asking your well spouse how their day at work was.



I'm not fully sure of what the underlying concept what is actually being measured is. I know Sarah keeps saying it is activity for this survey and there are claims of 'ME/CFS' (although there are a lot of behavioural reframings of physical function terms). But at best this seems to be a list of areas put down to assess 'where approx people might struggle' not to take account of the n/a or sum-total actual activity, as those who don't for example have any family would not be able to tick many of these but as this isn't eg an app showing 'daily activity' show that they just did different things instead if those things aren't on a list.

I'm a bit disturbed about how hidden the agenda is on this and what it is collecting.

I'm not sure any other serious illness charity would, as a main measure it is being suggested most will the illness would have eg in a clinic, accept a survey asking about intimate relations and using the term 'avoid' for activities that relate to physical function.

And I don't understand why someone would do it and think that they wouldn't first need to offer very good explanation for doing so.

And I don't find the suggestion of 'Rasch analysis.... you wouldn't understand it' and some talking about a favourite book a fair explanation at all.
 
Last edited:
I had thought it might be factor analysis - don't know where I picked that up from. Stats are not my thing, however from experience with kids it's common to run things through software without being aware of what it means so long as you get the answer.

https://www.theanalysisfactor.com/t...cipal-component-analysis-and-factor-analysis/
There are many different methods - is there a stats plan ?

Given a previous response was to list a huge variety of statistical tests and no method , it's pretty certain that another ship can dock at the previously charted destination soon .

Wine.. bottles ....

ETA I have not yet responded on FB so will try and do that tonight.
 
You must log in or register to reply here.
Back
Top Bottom