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

Really?

Doesn't sound the type of healthcare anyone would want for any other illness? She's hardly forwarded the argument she has in mind when she says this well - how can these be used to help someone's health?


And no, really no proper scientific medicine would be using any of these types of stats tools layered on each other - particularly when such poor checking/regulation of the input data has occured, this hasn't operated like even an experiment where they've drawn x number of people meeting y criteria in and asked them to do z.

They've used MEA to distribute extremely long surveys many won't be able to complete without detriment to their accuracy, and with many ambiguously-written questions that the intention of how they will be used have been hidden ( a massive informed consent issue) to use so many stats it looks like a fishing expedition on an apparently endless and not either verified or confirmed from the list within the new guideline (to ensure they are 'on task') load of features.

SO we neither know the symptoms/aspects being measured, nor the weighting of each, nor who on the people they've recruited. Which makes the quoting of means and std deviations laughable - what do they think those mean? deviation from the middle of what 'group' on what 'factor'? You have to at least be very sure on one (either the group or the factor and how it operates) for that deviation to means something because it is a check of eg representativeness or range of the then dependent factor. ?



The problem is that even if we did finally get some decent, normal, clever independent physicians who saw a pwme and didn't have a bias so just looked into this with an open mind

Then somehow rather than this sort of thing reflect on those who designed it, and then empathising with us on the type of thing we've had to put up with and weather for all these years, hence asking for/stating a complete change of specialism being 'in charge' is necessary...

somehow this ends up reflecting on us as if we must have done something to deserve it, or that we had any power in asking for this.

SO it is important for our reputation as pwme and how we are ever to be seen and respected by other staff and scientists going forward that MEA reject this. And say it 'isn't ours [pwme] as an invention'

or as if our condition is this complicated - when it mightn't be the most straightforward illness but actually if you take out the having to pander to a political compromise/needs of staff it at least means you are just talking about the actual illness not having to include some 'it might also be a delusion' aspect etc.

what we need is for us to filter out other people's wishes and reframing based on their wanting to keep their jobs as they were or not wanting to have to set history straight.

So we can just say it is a condition of PEM/PESE (and deterioration if we overdo it) along with x symptoms that we know are pertinent. ANd therefore comes in different severities which have vastly different debility patterns - but that these link on a spectrum because that 'environment' and deterioration means 'within individuals' can move on that spectrum. Just like with other illnesses where you go through stages and get worse.

We need something looking to keep the nuance and the parts that are important by severity, but strip out the political compromise stuff. The worst of all is trying to hide that, make us 'mysterious' by their working and so even if it creates more work we can't strip out the stuff that's the old school 'but it might be how they think' additions because they've embedded it into all the other measures by using this weird method and strange 'psychometrics' instead of basic, straightforward data.

The most basic right we all have after everything we've been through as a group should be access to our own data - to be able to check its vericacy and know its source. Given how many errors and updates to the concepts there have been in the past. I can't see how this is useful or interrogable?

 

I don’t think it’s anything “worse” than setting up a framework which will generate positive outcomes for NHS commissioners. But it could be more sinister I suppose. It comes across as “the tail wagging the dog”

The fact that nobody can give a concise, relevant, meaningful and substantial answer about this project is a big red flag. As is Sarah Tyson’s conduct. We’ve had insults, obfuscation and the dismissal of wearables as an esoteric unreliable something which the NHS cannot comprehend - despite other parts of the NHS using them.

 

To be fair, I'm not sure she gets the statistical tests she is citing - so I wouldn't assume that tone on that part is due to her assuming we know less than she does but that she perhaps is being passed on the info etc because there is a stats expert in her team/it's not her specific role.

 

It’s all part of a pattern of very defensive reactions though. Look at the tone without that part!

 

It is worse. Because the norm with these sorts of things is that you are coming up with 10 measures, and everyone who needs to be around a table then scrutinises these as measures and their weightings and methods - I've been part of how these things are done for other things.

There is a reason it is done transparently.

You then also get 'raw data' per quesitn, and you don't have to imagine (but not know) that because there are 50 other questions when people started not caring or not spotting the error in the question means they misread it etc. And can at least argue on x point if they then change something because 'there is no movement in question 9 from before and after treatment in 90% of attendees' because you have the raw data for each question and can see each question.

This strange methodology has made the data uninterrogable, unchangeable and no idea where the weightings are. For all we know on a black box algorithm it could be that only 3 questions have strong enough weighting and chance of variability due to ceiling/floor effects on their scales that the rest are just window dressing.

That's easily done - as we've seen with the critique of the CFQ scale, you can't get more exhausted than 5 and start on 5 because they've calibrated it wrongly for ME/CFS and so only those who don't have ME/CFS but have CF/something that won't get worse but might 'improve over 6months' will have changeable scores and therefore their 'movement' ends up having more impact that anyone else's as noone else's can move (due to the maximum being so low).

 

Yeah, someone might want to point her to the thread on EFfRT, or maybe that's why

 

What that?

 

Nope

I think the two important bits of what she did say to focus on are:

So some of us might want to be looking at whether we can get radical rest in.

The sad truth is that whilst the NIH at least did provide the raw data

I'm not sure the same standards apply here - which given the context (why isn't that a norm for this country, and then there is the fact it is a project funded by a charity for the illness, provided access to the participants and the 'trust' on which they filled it in despite the really worrying consent issues, and is about those with the illness) is a very important discussion issue in itself

 

It has a lot of people discussing the correct use of statistics and being able to analyse these things etc.

 

All the more reason you’d expect Sarah to explain in succinctly in plain language. Rather than a list of expert jargon and a whole sentence explaining how they did something not here, but on a different project?

I think PEM is separated because it can’t be “fixed”