Open The Chronic Illness Survey Adventure (Symptom Cluster Characterization in Complex Chronic Disease)

I'm concerned for conclusions such as 'hEDS patients and ME/CFS show high symptoms overlap so perhaps they are related or the same illness' or 'POTS patients show a lot of gut issues just like MCAS patients suggesting a shared pathology, something to look into further' etc etc...

Because of the way the study is set up (without sufficient care for selection bias, and without scrutiny of the validity of POTS, MCAS en hEDS diagnoses) I think such results are likely to be misleading.

Because this is a study done directly by one of the major patient organisations, I suspect that the study results will be referred to a lot and that it may determine future advocacy efforts. So that's why I wanted to raise my concern about this in time.

 

My first thought was how are they going to deal with the high misdiagnosis rate, wouldn't that be messing up the data big time, and extra double big time when you add the other diagnoses.

On reflection, there could be an opportunity here if it's done well. It looks like they're going to ask about a very long list of symptoms. Presumably they'll also ask what criteria your doctor used. And a subgroup will be coming from the clinics of selected ME specialists.

So that creates the opportunity to compare what criteria doctors used, if any, and what symptoms patients actually have, in other words do they meet the criteria their doctors used, some other criteria or even none at all. Also, how do patients diagnosed by the selected ME specialists differ from patients diagnosed by their GPs or other doctors.

I'm lukewarm on adding all the other diagnoses to the mix but I guess something may come from it if they focus on symptoms and don't get hung up on the current labels for clusters of symptoms.

The devil will be in the detail.

 

One a more positive note: it may be that diagnosis of POTS, MCAS, hEDS, ME/CFS, fibromyalgia, chronic lyme diagnoses all depends not so much on the patient's symptoms but on the doctors they have access to. In that case, it makes sense to group them together and look at the clinical characteristics to see if anything stands out for those groups merged together.

Also: despite my criticism of some aspects of the study design, I still encourage people to participate. If a really big sample size is reached interesting results may come out of it. Participation usually takes a relatively short moment of our time/energy but could help researchers further. And you'll never what might come out of it, always a small chance that something significant is found.

 

And why is this a problem? It's perfectly possible that for example hEDS is often just ME/CFS that was diagnosed differently because the patient happened to see a doctor that believed more in hEDS than ME/CFS.

We should be trying to find out if this could be the case or not.

The diagnoses given to patients are influenced by their own biases in society and the healthcare system. Not segregating patients by their diagnosis is a way to reduce the influence of these biases on a study. Of course the study design brings with it its own biases but maybe doing things differently will allow new things to be found which can be validated in subsequent studies.

 

It looks like they are going to recruit through ME/CFS patients organisations and ME/CFS clinicians so the results are probably going to be biased in that way.

In my view, looking at symptom clustering or overlap with a survey also isn't very reliable to understand a relationship between diagnoses. I remember a recent study where the authors argued that ME/CFS = fibromyalgia because both patient groups report similar symptoms. It was also based on long lists of symptoms that some argued that CCI and ME/CFS are related (when looking at descriptions of the clinical picture I got a very different impression, in other words, the symptom overlap was misleading).

 

I know what study you mean. I cannot remember the details but remember thinking (I hope I'm not remembering wrong) the study came to this conclusion because it measured a limited of common symptoms. It didn't allow patients to openly describe all their symptoms. It seems almost inevitable to arrive at similar symptom profiles with this methods.

 

On the website of the survey Massachussetts ME/CFS & FM is no longer listed as partner, but OMF has been added as partner.

 

"The Chronic Illness Survey Adventure"

Using the word "Adventure" in this makes it sound like it's fun to have a chronic illness. I think it's insulting.

 

Yes, my immediate reaction was that "adventure" is misplaced in this context.

 

I absolutely agree, @Michiel Tack. I am afraid I think this is entirely the wrong way to do things. It will generate exactly the sort of bias and noise that has worried me about DecodeME. It worries me that 'Edinburgh' is mentioned here.

Medicine does not work by making lists of symptoms and similarities between illnesses. That is a bit like looking at lists of words in newspaper headlines and similarities - you end up with something that has absolutely nothing to do with anything of interest.

To call this an adventure and to say that it is all about shouting things out seems to me to indicate what it is really about. It will simply generate more of the sort of noise that we see in so many useless publications these days. On the social scene I suspect it will simply encourage the average person in the street to say 'uh, oh, all the head I suppose'.

Nul points, I am afraid.

 

It feels like each individual's symptoms should be recorded without reference to their diagnosis. Any clustering of symptoms would then become evident, and only after that then see how those clusters do or don't map onto people's diagnosis. I appreciate that self-selection, non-representative sampling, etc. are separate issues.

 

Adventure term in research means ‘who knows where it’s going to get us’- basically a game of Survivor? The Great Adventure race? A bad case of reality TV?

 

I understand this perspective and it is clearly popular and dear to many but this is exactly why medicine has failed us. Medicine works from a disease process and when it can't figure it out the work just stops. I would definitely not count that as a valid argument. Especially given what limited understanding this has given us, disease doesn't explain most symptoms, everything is just associated. This isn't understanding it's just mindlessly cataloguing.

When the standard way of working fails, the only information we have is illness, which means symptoms. Trying the same old approach will get us nowhere until a major breakthrough happens on figuring out the disease process. Currently medicine is only looking in familiar places because this is where the street lamp is shining. That will not work. It is currently not working, locked in the same cycle of failure.

Medicine has reached the limits of what its traditional approach can do in most areas. In the last few decades only advanced biomedical research has produced meaningful progress and yet the "common wisdom" right now is to abandon science and go full BPS with EBM, a guaranteed failure that already has decades of disastrous outcomes. So raising the fact that this is different from the way that has failed is frankly an argument for, rather than against.

Seeing the abysmal quality of most of the work happening on Long Covid makes that point clearer than anything could, the system has reached its currently limits many years ago. The point of medicine isn't to deal with floating diseases, it's about people, that means illness and that goes through symptoms, the only reliable information we have. This is the paradigm shift that is needed. It's scary because is new but all we have otherwise is familiar failure and that's just the most possibly broken thing in the damn known universe.

 

Signed up, out of curiosity as much as anything, and have just completed the first part of the survey (they're sent out in monthly installments).

There were a few technical hitches:

When you tick 'other (please describe)' there isn't a field where to do the describing in.

I selected the non-gamified version and then received a couple of different-looking emails both stating rather confusingly "Some survey-takers received a link with broken code early on in the survey-taking process. If you received this email more than once, this email is the one that contains the corrected URL."

When you click on the back arrow once you're a few pages in it takes you right back to the very first page, not to the one just before the one you're on.

Also be aware that once you select an answer option (of the circle type) you can no longer change your mind should you wish to skip the question without answering after all. You can select a different option but you can no longer select no option, i.e. deselect all.

Regarding the survey questions:

They use some of the standard questionnaires about health and functioning, with all the associated problems of interpretation we've discussed on the forum elsewhere, like asking about the last 4 weeks compared to usual (what's usual?), or how you rate your health in general without specifying if that means apart from ME (pretty good) or with ME (awful), or function scales that mix up different aspects like how much daily assistance you need (a lot) and what level of medical care you need (not much).

They ask about a wide range of conditions diagnosed by a medical professional. What qualifies as a medical professional isn't specified.

There's no room to add comments about diagnoses received or not received. Apart from screening healthy controls for their actual level of healthy, it's not clear what the answers will be used for. This makes it difficult to answer meaningfully. For example:

I've been diagnosed with MS, and later was undiagnosed again. Is that a yes or a no?
I've been diagnosed with IBS and still am even though I have never ever actually met any diagnostic criteria for IBS. So strictly speaking a yes, but a yes would be misleading.
And the opposite, I very clearly meet diagnostic criteria for POT(S) but have never been diagnosed with it. So technically speaking a no, but a no would be misleading.

It may be that the intention is to look at how reported symptoms match reported diagnoses to get a handle on rates of misdiagnosis, which would make some sense, but that's not clear from this first survey.

Other questions are also too open to interpretation, e.g. they ask about being worse after a lumbar puncture. Which I was - I had nerve pain down my leg for a couple of years because of the needle hitting the wrong spot - but that's probably not the sort of worse they mean.

All in all underwhelming so far. Hope it gets better.

I would have preferred to get all the questions at once to get an overview. That often helps me answer questions more meaningfully. I know they're splitting the survey so as not to overload patients but I don't see what difference that makes since you have a year to answer everything and can save as you go and come back to it.

ETA:

Some of the diagnoses asked about are exclusionary dx in some of the ME/CFS diagnostic criteria unless well controlled.

So if one of the purposes of asking about such diagnoses is to confirm/strengthen the self-reported ME/CFS diagnosis then a follow-up question is needed: is the condition you've been diagnosed with (in addition to ME/CFS) treated and well controlled?

If the only purpose is solely to see what sort of other diagnoses cluster with ME/CFS such a follow-up question would be less necessary but in that case there should be a free text field to add any other dx they didn't think to ask about.

 

I do not remember reading that. But that question right there is a little biased because not everyone answering the questionnaire will have had the ‘chance’ to even get a lumbar puncture. There are strict guidelines to be ‘eligible’ for one, so even if i wanted one, i would not be able to have one over here. It sounds a bit ‘fishing expedition’ to me.

 

The process as described on the website:

"Fill out a quick contact form.

Register your interest by filling out the contact form below.


Choose a time to touch base.

Next, you’ll receive an invitation to chat with our Director of Scientific and Medical Outreach for a few minutes to receive login credentials. You pick the time that works for you.


Click on the link your patient sends you.

Click on the link your patient sends you and validate their diagnosis.

From then on, you can validate any patient’s diagnosis with one click."

 

I think you may have missed the point.
Symptoms are absolutely where medicine starts. My development of rituximab for RA was absolutely dependent on listening to symptoms - but DETAILS of symptoms.

Symptoms are only any use as part of a history. If we ask if someone has pain we get nowhere unless we ask six further questions that give us some idea what to make of it - how long it has been there, where is it, what is its character, what is its time course, what makes it better or makes it worse, are there any associated features like swelling?

The point is that there is no point in telling your garage man that there is a funny noise. He wants to know how long it has been there, where it is coming from, how often it happens, what brings it on etc etc. We understand symptoms if we have a certain minimum degree of detail. Without that there is no point. The lists I see for LongCovid and now it seems for everything, including Parkinson's disease, are really not interesting.

There is also no particular point in getting numbers on how common symptoms are. Splenomegaly is a rather uncommon feature of RA but is a key part of our understanding of what is going on.

To put it another way, I think I was in a position to make something of the features of RA and build an understanding of the disease on it after I had seen several hundred cases over a period of ten years, maybe 15. I had tried to do a cluster analysis five years earlier and because I was not attending to detail got nowhere. Many of the discussions on this forum go in to the sort of detail we need. I just don't think lists are going to be helpful.