Can subjective measures be reliable in research with pwME without safeguards/codes regarding e.g. over-exertion, coercibility etc?

I want to bring this up from a methodology perspective which I think is fundamental to studies into ME/CFS past and present. Basically terming, rightly in the context of studies, pwme as 'vulnerable' under certain conditions and flagging that data taken under these without safety adjustments can surely not be usable.

I would like it to become a big topic for research and certainly know e.g. scientific psychology (and likely other scientific-based subjects) to discuss how 'conditions' can make answers or measures effectively invalid. List what the risks are, and how these could/should be creatively begun to be addressed and indeed for research that heeds none of these to be binned on that basis of unreliability and consent issues.

Due to our vulnerability regarding energy, and this affecting very clearly from a scientific psychology onlooker's perspective the ability to 'fight' in a conversation where someone is being 'assertive', or even to summarise and really feel confident of thinking or reading correctly once way past energy threshold.

I personally have concerns that any research that has relied only on measures that are not objective need to be thrown in the bin unless they have within their methodology good explanations about how these responses are both with consent and when someone was in a position to feel confident of them (and not being 'led').

The topic of how any of these measures can be ensured to be accurate, consenting and without coercion, influence or detrimental state I think needs to be a prime topic for ME/CFS. I know if I e.g. ran a focus group/wanted to work out how to get input from those with severe ME I would need to think so cleverly and creatively to come up with a method whereby I knew people were able to contribute to their own satisfaction (and it wouldn't be just 'running a 45min focus group online' expecting top of the head answers they can't amend).

Would for example someone forced to give feedback or agree to something at that point in time the measure was taken still, if allowed to amend their answer a fortnight later when out of PEM and exertion and in a safe place, still give the same answer? Why is this protection not part of all research designs for pwme?

It certainly feels inappropriate that any pwme should be asked to complete any 'measure' after attending an appointment, or even just a journey somewhere due to the exertion before. Add in a clock to complete something, or just the internal one as people are over-tired and desperate to get home and you've got theoretically coercion/not correct circumstances for fully considered answers issues.

There are many other component parts to this and why I do think that this area of methodology needs to be taken very seriously indeed as not reliable without proper design being researched and I'll try and put them down as I remember but would appreciate others contributing ideas and their own thoughts?

 

I know certain personality types of 'communication techniques' can basically outpace our conversation/thinking threshold and force/walk us to say things we don't agree with or mean. And hours afterwards even the most assertive of us would be stunned at how someone has managed to do that. But when you are beyond your cognitive or energy limit and someone is doing 'foot on neck' interruption and cutting you short techniques you are being controlled and not allowed to escape to safety until you have said what they want.

I know most research designs involve exertion prior to filling out any questions which makes pwme certain to be exhausted at that point, and I never see in the discussion how the questionnaire was designed to ensure it was within the cognitive limits of the individual at the time. A lot of research involves numerous questionnaires. In fact I was shocked at one maybe 18months back that I counted over a 100 questions for - which I think were done numerous times over. This is surely 'pushing on the vulnerability' and completely unnecessary - no research requires that many questions if it knows what it is doing or looking for and not 'fishing'.

Then there is the power differential and people-pleasing issue with all sorts of potential perceived 'risks' from displeasing, the risk of letters to someone else suggesting 'lack of effort' and the impact of that being one.

Should one standard be that all ME/CFS research should provide a right of amendment to answers a fortnight later which subjects have a right to do in their chosen environment in order that they are not in exertion?

And another be a board of some sort that assesses the cognitive 'load' and 'tricks/pitfalls' on questionnaires to ensure that tools are neither individually likely to be not appropriate to the disability of pwme as subjects (poor wording, questions that don't make sense) as well as the length and repetition being suitable

- I think clarity of the purpose of any research in order that such questions are contextualised and made clear is also another vital one, few areas are so fuzzy and misleading historically so this adds a further load to questions. But also it is fair to acknowledge how so much research has misled and issues of informed consent - a basic right needing to be firmed up in proper methodological requirements that are fit for pwme and to the context of ME/CFS research.

 

Ok, maybe the term I'm looking for is 'a code of conduct' for methodology, but one strong enough that it goes to ethics committees as something serious highlighting that where methodologies fail to account for these vulnerabilities in design there will be questions regarding reliability and validity as well as ethics.

And maybe, whilst some issues are ME-specific, this is something that could garner other conditions which are energy-limiting or have other elements that makes for vulnerabilities?

 

And yes, I do believe it should be straightforward to do studies showing exactly how responses can be influenced/changed by use of certain techniques for pwme.

I've long thought that the CBT-related ones, and their various measures - before we even add in the conditions like people being exhausted when filling it out - are becoming increasingly popular because they are like slot-machines in always giving out the 10% if you have people with certain conditions certainly.

I do think that research showing controls using BPS methods/conditions replicated and then a 'vulnerability protective' measure and process should hopefully show up the difference. There are probably all sorts of other designs which could demonstrate how pwme can be 'walked into giving x% response' by certain designs and conditions.

And I think that this is important because given the lack of blinded controls then basically quantifying the 'placebo' that this is used as by employing certain techniques it could then be 'taken off' the end 'effect' or significance level in order to see and estimate how much is left after that accounted for by vulnerability and leading effects.

I mean these could be in ridiculous things like getting people to say they like a certain football team or pop band more than they did at the end (being more positive about it) but demonstrating it in simple papers that laypersons and those preferred storytelling type articles and books could use well I do think it is time these research things were done in order to communicate what might be happenning in a simple way.

 

Indeed it could begin with simple surveys that give pwme a list of different question types (leading, worded with 'not' the wrong way, using scales that are hard to understand what they are getting at)

and a list of different circumstances/conditions e.g. when completely rested, have done a journey, have already had 10mins of reading or talking and so on, needed to get home to bed, were in a people-pleasing situation or pushy conversation, short clear questionnaire they'd been given the context of clearly before vs long 20 question rambling one that didn't seem relevant.

and note whether they would be happy or confident that their answer was reliable or meant anything, or more precisely to red flag areas where they would have felt that they were 'not in control' of their answers to the point of them being invalid.

Could this give a jump-start to flagging the issue in order that funding/interest be obtained from methodologists who then might want to look into the limitations of pwme vs research designs that work and how problematic other types are?