Protocol Serial Paediatrics Omics Tracking in [ME] (SPOT-ME): protocol paper for a multidisciplinary, observational study..., 2024, Armstrong+

[MelbME]

The whole personality and behavioural field is far less sophisticated that it appears from a distance. I have been shocked at how ratings of serious mental health issues are derived from often inane questions. The survey instruments are claimed to have validity, but if you look deeper, the assessing of validity is often pretty spurious. The answers to the questions are scored, producing numbers, and suddenly the tool is no longer just a collection of ambiguous and often trivial questions. It is 'scientifically valid data' that can be used, context-free, to put labels on people.

There are all sorts of biases that can affect the data. Participants for a trial may be selected according to their attitudes; the participants that make it to the end of the trial and fill out the forms are often self-selecting. There is often priming, for example suggesting that people with ME/CFS are perfectionists, people will be encouraged to find the instances of trauma in their history, they will be encouraged to label themselves as Type A strivers, or people pleasers. And people will do that, because if there is a personality problem causing their illness, they can fix it, and become well.

Care of people with ME/CFS, even children, is not some magic thing that pediatricians have some special insight into. It is largely commonsense because there are no disease modifying treatments. I think you have enough skills and knowledge to look at the surveys they are asking the parents of the young people to fill out and at the treatment the pediatricians are recommending in order to know if what is happening is ethical or not.

That is indeed shocking, to think a few questions could be a diagnosis. The lack of physiological knowledge in this disease has lowered the bar for evidence to be actionable and it's not the only disease like that. I'm not a clinician and I expect their job is very difficult, especially because they are used to treating, people with problems need treatment. I'm hopeful with younger clinicians this is changing, it's a complicated job and I do unfortunately think the quality of clinician varies widely.

What you're describing is bias that we would be looking to avoid in interpreting the data.

 

[MelbME]

I think it is irrelevant to the concern I expressed whether anxiety and depression are thought of as physiological or mental health diseases.

The concern I expressed was with the choice of the HADS survey tool and others. HADS has been found to over-diagnose anxiety and depression in people with chronic debilitating illnesses. If you look at the questions in it, you will see why. It is not a good tool to assess anxiety and depression in ME/CFS. I'll add the link to the HADs thread again.

My subsequent question in fact showed that I believe that mood disorders are likely to have physiological causes and/or consequences. I think you missed it? I asked, if you are wanting to correlate mood with your omics, would it not be better to collect simple and direct reports of mood from the young people at the same frequency as your collection of samples? It seems to me that a very long questionnaire completed by the parent about the young person very infrequently and almost certainly biased by their beliefs around the cause of ME/CFS are unlikely to capture the detail you need to find good correlations.

I myself have been assessed by HADS a couple times, it never led to a diagnosis of anything. I do find it troubling that others have found different experiences. From clinician I speak to a survey might be one small part of a diagnosis, maybe even just a screening tool to put someone through a much more rigorous diagnosis process.

Yeah HADS is very broad, anyone suffering a chronic disease would answer higher.

 

[MelbME]

As I said, I don't think the ethics committees would have insisted on full surveys being applied - we see sub scales used on their own all the time.

But, all this raises the question we have discussed on another thread: 'should studies be being done on children with ME/CFS at this time at all?'. This study could have been done with adults, and presumably it was only done on children because the neuropsychologists based at the Royal Children's wanted to get hold of data that would help support their causal paradigm and treatment ideas. Actually, given the turbulent time that adolescence is biochemically, I think a strong case could be made for the omics study to not be made in this age group. There will be a whole lot of biochemical noise from puberty muddying the waters.

The risks that flow from the application of the particular surveys chosen are real.

Yes it could have been done on adults but we felt there was little biological research on pediatrics, felt it was a section of the community being under served and wanted to improve that. That was the goal of the NHMRC grant we got.

 

[MelbME]

Thank you, I really appreciate you replying here.

Let's not get sidetracked by questions about whether people with ME/CFS have a fear of being associated with anxiety or depression; Or indeed whether my son and I had a negative experience with the clinicians. I hope you will have time to get to the bottom of what happened in this study.

I think the most important question we need a clear answer to is 'were both the young people and their parents specifically informed that the parents would be completing surveys about their children in order to identify psychopathologies, before they gave their consent to participate?

It occurs to me that the inclusion of the neuropsychology evaluations under the umbrella of the omics study is likely to lead to a bias in who sticks with the study and who drops out. Certainly, if I and my son had signed up to the study and then found out that I was expected to answer questions to evaluate his aggressiveness, his hyperactivity, the presence of a conduct disorder, his somaticisation, his withdrawal, or to assess his social skills at a time when he was not able to get out much, I expect we would have withdrawn from the study. Not because my son is in any way psychologically unhealthy, but because I would be concerned that the impact of the disease would bias the results and about the intent of the researchers. And frankly, it's none of the researchers' business.

And, if you have all the people affronted by the personality survey dropping out, and leaving the parents who do think that their child's ME/CFS is caused by their psychological issues, then you end up with both the neuropsychology arm and the omics arm in a great biased mess.

Yes the participants and guardians are provided with the full information of what will be performed during the study prior to their consent.and they can withdraw whenever they want, they could withdraw their data now if they wanted and we would delete it all.

 

[MelbME]

Unfortunately, there is no good answer to what happens with the results of the personality/behaviour surveys. If the results, skewed as they will inevitably be and almost certainly over-diagnosing somaticisation and withdrawal, are sent to the young person's GP, the results could greatly affect the care that the child is given and the way both child and parent are treated in the medical system well into the future.


I'm still intrigued to know what you mean by complex.

"Complex" term I'm using is intended to be short for complicated, multiple factors and heterogenous.

 

[Hutan]

Thanks for your replies @MelbME and your obviously good intent.

 

[Trish]

Hi @MelbME. Thank you for your continuing engagement here and your willingness to listen to criticism and suggestions. All too rare among researchers, we find.

Have I got this right? As a non clinician, the only way you could access a cohort of child patients for your omics study was through a group of clinicians who diagnose and treat children with ME/CFS. In order for them to agree to be involved in the study and recruit patients for you, they decided they would do some research of their own using questionnaires.

You are therefore stuck with your cohort of participants being concurrently researched on omics and on psychobehavioural factors.

My question is, do you, in order to fulfil your ethically approved protocol, have to include analysis of the psychobehavioural questionnaires alongside the omics data? Or can you simply ignore the questionnaire data and publish an omics paper?

To me it seems infinitely preferable that you publish the omics data without contaminating it with doubtless spurious associations with psychobehavioural stuff.

 

[Utsikt]

But also scales are very useful for comparisons across disease in this group and especially for assessing social impact. Even if a paediatric patient with ME/CFS were to get most of their physical function back they can still be quite impacted by having the disease at a younger age, as you might expect.

Are the scales calibrated or verified with regards to the unique characteristics of PEM? There’s a reason FUNCAP was created - PEM does not fit with most models of how disease and the body works.

This is a collaboration between us and neuropsych researchers that work with children. Their expertise put together the decision on scales to use as it pertains to neurocognitive testing, behaviour tests and mood tests.

This is not very encouraging. The research standards for both the neuro and psych fields are rather poor, and I can’t imagine that it gets any better if you combine them.

Are the scales «good enough» for proper research, or are they just «what people use». The latter does not guarantee the former. Neither does an ethics clearance. You have an independent responsibility to ensure that any measurement tool in your study is good enough, that can’t be outsourced.

I guess I would say that information or questions aren't biased, they are just data.

Data can have bias that was introduced during the measuring phase. The tools that collect the data can also have bias. Data isn’t some infallible our neutral thing - it’s quite the opposite.

 

[rvallee]

I am interested to know your perspective on diseases like anxiety and depression. Do you see them as physiological diseases or as separate mental health diseases?

The bigger problem is that they can't be reliably identified, so we can never know what is actually happening. There are no tests or markers for either, they are at the very best a guess. When the standard description of anxiety is worrying about outcomes, but then it's been turned into having symptoms of sympathetic arousal resembling stress, or whatever, and people routinely get labeled with either or both after infectious diseases or other causes where the whole process isn't understood. They are generically valid concepts, but for individual cases they are far less accurate and valid than even a chronic fatigue description. The state of knowledge is comparable to medicine's knowledge of autoimmune diseases in the 17th century. At least I haven't seen anything better than this.

I only ask because I sometimes get the impression that ME/CFS patients don't want to be associated with co-morbid depression or anxiety because they don't see those diseases as physiological. If I were doing research on those diseases I'd be looking into the physical/biological changes that happen around neurotransmitter receptors, metabolism, etc.

Definitely. None of this relates to my experience in any way, and the confusion I described above, along with the decades of miserable failure, means I want absolutely no research dealing with this stuff. Ever. It's simply not useful and never will be, because no one knows what these things are or can differentiate them.

Most likely most cases have a physical cause that can be identified with more advanced technology, but that's about as useful as sending a radio signal to someone who knows nothing about electromagnetic radiation, antennas or anything related to electrical circuits, and definitely doesn't have a radio or anything more complex than using a coconut as a cup. It will be decades before medicine does something useful here, and mostly likely it will be AI that does it. This stuff is just too complex to deal with for now, human nature just gets in the way and no good can come out of it for us.

It would be useful to understand this, but psychosocial questionnaires ain't the way. It just categorically isn't. Most of them are plain weird and reflect deeply disturbing beliefs held by people who clearly struggle to relate to lived experience different than their own.

 

[Hutan]

You have an independent responsibility to ensure that any measurement tool in your study is good enough, that can’t be outsourced.

@MelbME, it would be helpful to know if you have overall responsibility for the combined project, and for ensuring the participants and their families are not harmed by the project. If there are in fact two separate projects, who has responsibility for the neuropsych project?

Yes the participants and guardians are provided with the full information of what will be performed during the study prior to their consent

It would also be very helpful if the patient information sheet (and any other document that prospective participants were given to explain what the project involved) could be posted here.

It seems possible that either the young people or the parents or both did not understand that the parent would be assessing their child using tools for identifying psychopathology, and providing that information to the researchers. And that information would then be passed on to the young person's GP (with or without the parent and the young person being aware of the findings - I'm not clear on that yet). It seems possible to me because I can't imagine signing my child up for the study under those conditions, and I'd be surprised if many parents, especially after having dealt with the stigma of ME/CFS from medical professionals, would do that.

If that is indeed what happened, that participants were not completely informed, I think it would be good if they are told retrospectively, and told that they can participate in the project without participating in the neuropsych evaluation, and were reminded that they can withdraw data.

 

[rvallee]

The paediatricians have their totally-validated-you-guys questionnaires and they and ethics committees will very likely insist on those in standard form. I guess it will continue to be GIGO until we can ultimately show what's actually going on physiologically. Then there's a chance those psych instruments might be corrected or reduced to reflect the real situation and finally let go of the closely held somatisation etc.

I think we can safely use the example of peptic ulcers here to answer categorically that none of this will be relevant. In fact in all the years of being here every day on this forum, reading through hundreds of papers, most of which are about pushing a particular psychosomatic school of thought, peptic ulcers have simply not been mentioned once in all this time. There is simply zero consideration for any psychosocial factors in peptic ulcers anymore, even though it used to be a significant industry in its own right.

Which is all that's needed to know whether this stuff is useful. One day there will be a breakthrough pointing us to the right biological process/mechanism, and from this day literally everything psychosocial will become obsolete. All of it. And just the same for all the stuff that get mislabeled as 'anxiety' or 'depression', which are far more likely to be many causes leading to many different outcomes that simply happen to crossing in a few key places.

All this stuff is only considered relevant here because we haven't had this tipping point. But we know for a fact what happens once the scientific process finds a beachhead, and it doesn't bother with this stuff because it produces absolutely nothing of value or interest.

 

[rvallee]

Are the scales calibrated or verified with regards to the unique characteristics of PEM? There’s a reason FUNCAP was created - PEM does not fit with most models of how disease and the body works.

The deeper problem is that it's simply not possible to do this. There is no 'calibration' here in the typical sense that people understand, like the way weights or lengths are calibrated using reference units, which themselves are precisely measured.

Biased things are being compared to other biased things using some form of interpretation, this is simply a completely different approach to how, say, temperature scales were calibrated using physical experimentation involving very precise measurements.

IMO questionnaires asking non-objective (i.e. your height, your age) questions simply cannot be calibrated to a scientific level, not anymore than it's possible to create a mathematical formula to categorize types of poetry or whether a policy is fiscally liberal or conservative, they simply depend on value judgments whose internal metrics are unique to each subject.

The original sin was described in a comment above, this silly idea clinical psychologists (and psychiatrists) made up to think that turning subjective terms into relative numbers on non-linear scales makes them precise. This is simply not how science works.

 

[Hutan]

"Complex" term I'm using is intended to be short for complicated, multiple factors and heterogenous.

But, we don't know that ME/CFS is complicated, it may be simple, it's just that we don't understand it yet. What specifically is complicated? Wouldn't 'poorly understood' be more accurate if that is what you mean?

What multiple factors are involved in ME/CFS? Do you mean factors relating to onset? What factors are those? I don't think we know those specifically yet. If you mean that there is an interplay between host factors (some of them produced by genetics) and a pathogen trigger, isn't that true of every infectious disease? I don't think we label the common cold 'complex' just because that is true.

Heterogenous - do you mean the presentation is varied? That is certainly true, but even the common cold is expressed differently by different people, but, as I said, I don't think we typically describe the common cold as 'complex'. Or do you mean that among the people diagnosed with ME/CFS there may be subsets of two or more different diseases? That might be true, but we don't know that yet. In any case, 'complex' does not sound like a very useful term to get that across in the meantime. Why not just use 'with a heterogenous presentation'?

'Complex' is a vague word, liable to misinterpretation. It is often understood in medicine to mean that psychological factors are contributing to the disease. Given all that, and the presence of other words that I believe could convey your meaning better, isn't it best that that word is avoided when describing ME/CFS?

 

[Utsikt]

The deeper problem is that it's simply not possible to do this. There is no 'calibration' here in the typical sense that people understand, like the way weights were calibrated using reference weights.

It’s still possible to make a reasonable judgement on how well a questionaire and its results matches with how the patient group understands and experiences their illness.

I’m not talking about physics levels of calibration. Calibration means something completely different in qualitative medicine than in quantitative medicine. Even FUNCAP is only calibrated to severity level.

I’m not looking for perfect, just better and closer to good enough.

 

[Hutan]

There is simply zero consideration for any psychosocial factors in peptic ulcers anymore, even though it used to be a significant industry in its own right.

I don't think that is true. There are still doctors trotting out 'stress' as a cause; I think that has been documented on the forum. There are still doctors pointing to psychosocial causes in multiple sclerosis. People generally love to say that people suffering misfortune have brought it on themselves due to their wrong behaviour and wrong personality, so that they can feel safer. Sometimes, these ideas might even have a grain of truth - perhaps surges of adrenalin do play some part in peptic ulcers, or the behaviour of smoking may increase the impact of the HP infection?

My point is, even with a clear biological cause found, some people will still blame those sick people over there for being sick. It's human nature. So, we need researchers, especially neuropsych researchers, to be really aware of that human bias.

Just adding a link to ME/CFS Skeptic's excellent blog series on peptic ulcers. The weird theories that were not only expounded but widely believed are incredible.

 

[Maat]

Any information from this study we did goes back to their doctor, that doctor before the study would be the one to make decisions, if any, like referrals.

On that child's medical notes for life: potentially influencing every subsequent diagnosis and treatment (or lack of it), they ever receive. Without the knowledge of the parents or the child going into adulthood. I find this truly shocking, if I'm understanding that correctly.

 

[Utsikt]

On that child's medical notes for life: potentially influencing every subsequent diagnosis and treatment (or lack of it), they ever receive. Without the knowledge of the parents or the child going into adulthood. I find this truly shocking, if I'm understanding that correctly.

What?!

They are planning on giving data from the study to the participant’s GP? What’s the purpose? Why and/or how would that benefit the study or the participants? Is this normal? Why is the GP involved at all? I’m very confused.

 

[Jonathan Edwards]

When can it be justified as a part of care?

I said 'might', being hypothetical.

 

[Jonathan Edwards]

The scales are standardised way to get information on behaviour and mood from this population. These standard scales allow you to compare to other diseases. As opposed to just asking any random question, which is fine if consistent, but you limit the ways you can discuss the data point and what it means if you don't use standard scales.

But none of that justifies their use in pure research, Chris. They are intrusive and involve phoney concepts like somatisation. Psychological intrusion can cause huge harm. It destroyed a whole branch of my family. Maybe in a care context sometimes it is helpful but unless a research project is offering something therapeutic I see no justification in involving such intrusion at all.

A similar situation arose in the context of a rituximab study that was designed to provide explanatory scientific information but would yield no useful information on the value of giving rituximab which we already knew could kill people. I made it clear that I thought the study was unethical. Scientific studies that just hope to provide some useful theories cannot justify things that we know can do clinical harm.

 

[Jonathan Edwards]

I have only now caught up with this thread, which raises a whole lot of issues. I guess that the psychobehavioural studies may just reflect the practice of the people looking after the children any way but I do see a very major danger in combining these with omics and ending up with papers that appear to validate concepts like somatisation.

Non-clinical researchers need to be aware just how dumb and interfering clinicians can be in this area. Maybe this will just be a lesson learned. But I would advise against trying to collaborate with anyone who believes in a concept of 'somatisation'. This is the incompetent psychology that has been so damaging for people with ME/CFS.