Establishing a consensus on ME/CFS exclusionary illnesses 2022, Jason et al

[Sly Saint]

ABSTRACT
Introduction
The diagnosis of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is generally reached based on meeting the criteria of a case definition and eliminating other conditions that could be causing the patient’s symptoms. Investigators in the ME/CFS field are currently confronted with 139 medical exclusionary conditions listed among the various ME/CFS case definitions. There is a need to standardize the illnesses/diagnoses that should be excluded.

Methods
Exclusionary conditions were listed for several prominent ME/CFS case definitions. From this list, symptoms were also identified as exclusionary by several physicians with experience in diagnosing ME/CFS. Input was also solicited from representatives from the patient community for a consensus list of exclusionary comorbid conditions.

Results
Once overlapping illnesses were eliminated, a consensus was reached on a briefer set of exclusionary conditions. The final set of exclusionary conditions is divided into 14 categories with 53 specific examples.

Conclusions
It is important for ME/CFS researchers to select uniform medical conditions to exclude from their studies so that samples across different studies are consistent and generate generalizable ME/CFS findings. This list can be applied to ME/CFS case definitions in order to enhance the reproducibility of identifying patients with ME/CFS for research studies.

https://www.tandfonline.com/doi/abs/10.1080/21641846.2022.2150487?src=&journalCode=rftg20

 

[Shadrach Loom]

Seems very sensible. But also potentially reminiscent of the XKCD cartoon about standards.

 

[Andy]

Great that patients had input, poor that it's behind a paywall.

 

[Samuel]

i am confused.

it probably all makes sense though. prob just me.


in the state of texas, there is a rumored needed killin' law. if there is a paywall, it needed removin'.


seid or something said somehting like it is a positive diagnosis, not a diag of exclusion. contradiciton.

the word comorbid has 16,000 definitions. without a definition, it is hard to interpret. m.e. causes disease a. or helps cause it. or a is unrelated. or disease a causes or triggers m.e. etc. which is it? whhy is it exculsionary?

m.e. is a @!#$ SERIOUS disease. it is multisystem, complex, and systemic. it attacks practically everything. yet the idea is that there is a coherence to it. ok.... let's buy that for a minute.

where do we get off knowing it isn't m.e. just because you have disease a ALSO? are researchers studyig severe? ok maybe we aren't knowing it. we are just positing it for hte sake of a research study. i can get behind that. except the other stuff i am saying.


otoh, if we present as classic b12 symptoms, if ther eis such a thing as classic, and it HAPPENS to overlap an m.e. set of criteria, does it?, then we can say after working up for b12 that it is exclusionary? is that the idea?


you say exclusionary disease a can cause all the symptoms described by the pwme in the 2.6 seconds allotted in the visit while the doctor is looking at the nurse's new uniform and thinking of the car he is going to buy and lunch.

while the pwme has oi and had to rest for a month to get to the appt and could have permanent harm. is the pwme going to list everything? or reember everything? couldn't that change the outcome?


where was i? m.e. can CAUSE all sorts of things. are we SURE it cannot also ause a bunch of other things? ok so you say those other things are a covering set for that pwithoutme so it isn't m.e. even so, could that be false? hickam vs. occam.

what about studying those whose m.e. has lasted for many many years and thus comorbidities by whatever definitsions? could the list chnge?

i'm sure there are pefectly respectable answers to my questions. he is talking about research definitions not just clinical. statndardiaztion can be good in principole. so yeah, let's eliminate outliers. just make sure 1] they are outliers, and 2] we circle around back to them.

i'd ust like in my brain not working state for this to make omre sense. i do not think this is obvious stuff. and when somthing is not obvious i think it needs careful wording, especially to get rid of ambiguities and be transparent about logic step by step if nec.

 

[Trish]

It's hard to judge the usefulness of this exercise without being able to see the whole paper, not just the abstract.

I think the word exclusionary can cause confusion.

Being diagnosed with ME/CFS should be a positive diagnosis of the presence of disabling fatigue, PEM, OI, cognitive and sleep probems (the list varying a bit depending on the definition used). This is obviously important for the patient clinically, so they get the right advice on activity management, and the right financial and personal support.

But some of these symptoms can also be caused by other things, eg fatigue by some nutritional deficiencies, OI caused by POTS etc. So there should be testing to see if these are present as well. In that context, exclusion of other causes is about ensuring treatable causes for some of the symptoms are found and treated. It doesn't mean you don't have ME/CFS. It's possible to have more than one condition causing overlapping symptoms.

For research, exclusionary can mean that if someone has both, for example, ME/CFS and thyroid hormone deficiency, even if treated, that may confuse the picture in the research results, so it may be necessary to exclude some people who have both ME/CFS and something else from the research so that any findings are likely to be caused by the ME/CFS, not the other condition.

 

[adambeyoncelowe]

The longer the list of exclusions is, the more chance people with comorbidities will be left out, that's true.

But this is for research (although that only becomes clear with the conclusion: "It is important for ME/CFS researchers to select uniform medical conditions...").

For research, you want the "purest" signal you can get, to limit confounding factors.

In real life, there will have to be some leeway, as plenty of people will have comorbidities -- even exclusionary conditions that might otherwise mean you wouldn't diagnose ME.

For instance, bipolar would probably be an exclusionary diagnosis. But I can see how someone with well treated bipolar who suddenly comes down with ME-like symptoms after a virus might get diagnosed with that as well.

It's possible they just have bipolar and the recent virus has affected their control of their illness, but it's also possible they have both.

So you need to watch and be willing to revise your diagnosis if necessary. If the patient suddenly recovers when their bipolar meds are adjusted, you might remove the ME diagnosis and put it down to misdiagnosis.

Conversely, if their bipolar is still really well managed, and yet ME symptoms persist, you might find yourself vindicated in giving that diagnosis.

 

[bobbler]

The longer the list of exclusions is, the more chance people with comorbidities will be left out, that's true.

But this is for research (although that only becomes clear with the conclusion: "It is important for ME/CFS researchers to select uniform medical conditions...").

For research, you want the "purest" signal you can get, to limit confounding factors.

In real life, there will have to be some leeway, as plenty of people will have comorbidities -- even exclusionary conditions that might otherwise mean you wouldn't diagnose ME.

For instance, bipolar would probably be an exclusionary diagnosis. But I can see how someone with well treated bipolar who suddenly comes down with ME-like symptoms after a virus might get diagnosed with that as well.

It's possible they just have bipolar and the recent virus has affected their control of their illness, but it's also possible they have both.

So you need to watch and be willing to revise your diagnosis if necessary. If the patient suddenly recovers when their bipolar meds are adjusted, you might remove the ME diagnosis and put it down to misdiagnosis.

Conversely, if their bipolar is still really well managed, and yet ME symptoms persist, you might find yourself vindicated in giving that diagnosis.

I'm divided on this even for research I'm afraid. We don't know enough about ME/CFS to know that there aren't certain types or just part of the spectrum that either have more misdiagnoses,'looks likes', or comorbidities that come with the ME that means we are defining the illness narrowly and leaving people out but also missing a key par of it.

I think we risk making the illness look like something only people who don't have certain symptoms have - the condition in itself may well come with some of these things, or they may be so likely long-term if you have ME and a propensity it's pretty important to work out which is chicken and which egg. There may well be connections that lead to what is underlying all this which are missed if you cut all those people with the clues out. Imagine if shingles was investigated ruling out those who'd had chicken pox to be extreme. Or autoimmune conditions ruled out other autoimmune conditions or certain prior illnesses - it would wreck understanding of these.

And most research is more like spectrometry where you have 'different peaks' for each different element, so actually the more inclusive you are the more you actually describe the illness in full as a good start. It does come with susceptibility to other things it may cause susceptibility to other things other things might make you more susceptible. Looking for this unicorn of only the 'purest' sounds like nonsense. You need big picture first.

I agree with proper checking on PEM - particularly to underline the issue with BPS still seemingly encouraging medics to 'think fatigue' - and for this criteria to be well-defined. But after that no, we need all and in big numbers. And the better the medical history they can provide the better - rather than those who might get something else in the future or just never got diagnosed with these supposed 'extras' being prioritised over not. That makes no sense and feels a fallacy to me.

Who are these people who got ME, nothing else and managed to never be misdiagnosed or diagnosed with anything else - and that doesn't mean it isn't because someone didn't look for it? We have no idea that is representative of what ME is at all until you actually get a good representative sample - and if 95% have something else what real use or mislead is recruiting from the 5%, without even checking if they have different make-ups from the rest that maybe just mean a common, normal element of the disease is suppressed.

All clusters need explaining in relation to each - indeed it might be that it is the 'least pure' that explains the others, or a point in the time sequence of what happens.

But more to the point the people you are actually excluding are those who know and can define what they do and don't have and what is due to what and have been investigated to know they still have ME. The rest haven't been investigated. SO the idea they are purer is nonsense. They are just question marks in truth

I also think given the state of what those who have had ME for a while in the UK currently will have been put through no diagnoses used for exclusion should be taken only from old notes - they have to be ones that the patient and current doctor are sure of, because so many in the past ended up being shoved into buckets that never made sense yet were never removed for so many reasons (including GPs KPIs, things never being removed from ntoes, not good descriptions of PEM meaning patients didn't have the lagnuage to explain symtpoms and felt desperate etc).

 

[Peter T]

I think the word exclusionary can cause confusion.

I agree with @Trish. Given ME is a positive diagnosis, surely this discussion shouid not about excluding diagnoses but establishing a system of ‘differential diagnoses’. It is not uncommon for people to have concurrent diagnosis, so it makes no sense to say for example someone can only have ME when the following have been eliminated.

An exclusion approach would say that someone could not have ME when they have for example thyroid issues or when they have B12 deficiency, however though it is not the norm it is not that rare for people to have ME and additional thyroid issues or ME and B12 deficiency.

 

[adambeyoncelowe]

I'm divided on this even for research I'm afraid. We don't know enough about ME/CFS to know that there aren't certain types or just part of the spectrum that either have more misdiagnoses,'looks likes', or comorbidities that come with the ME that means we are defining the illness narrowly and leaving people out but also missing a key par of it.

I think we risk making the illness look like something only people who don't have certain symptoms have - the condition in itself may well come with some of these things, or they may be so likely long-term if you have ME and a propensity it's pretty important to work out which is chicken and which egg. There may well be connections that lead to what is underlying all this which are missed if you cut all those people with the clues out. Imagine if shingles was investigated ruling out those who'd had chicken pox to be extreme. Or autoimmune conditions ruled out other autoimmune conditions or certain prior illnesses - it would wreck understanding of these.

And most research is more like spectrometry where you have 'different peaks' for each different element, so actually the more inclusive you are the more you actually describe the illness in full as a good start. It does come with susceptibility to other things it may cause susceptibility to other things other things might make you more susceptible. Looking for this unicorn of only the 'purest' sounds like nonsense. You need big picture first.

I agree with proper checking on PEM - particularly to underline the issue with BPS still seemingly encouraging medics to 'think fatigue' - and for this criteria to be well-defined. But after that no, we need all and in big numbers. And the better the medical history they can provide the better - rather than those who might get something else in the future or just never got diagnosed with these supposed 'extras' being prioritised over not. That makes no sense and feels a fallacy to me.

Who are these people who got ME, nothing else and managed to never be misdiagnosed or diagnosed with anything else - and that doesn't mean it isn't because someone didn't look for it? We have no idea that is representative of what ME is at all until you actually get a good representative sample - and if 95% have something else what real use or mislead is recruiting from the 5%, without even checking if they have different make-ups from the rest that maybe just mean a common, normal element of the disease is suppressed.

All clusters need explaining in relation to each - indeed it might be that it is the 'least pure' that explains the others, or a point in the time sequence of what happens.

But more to the point the people you are actually excluding are those who know and can define what they do and don't have and what is due to what and have been investigated to know they still have ME. The rest haven't been investigated. SO the idea they are purer is nonsense. They are just question marks in truth

I also think given the state of what those who have had ME for a while in the UK currently will have been put through no diagnoses used for exclusion should be taken only from old notes - they have to be ones that the patient and current doctor are sure of, because so many in the past ended up being shoved into buckets that never made sense yet were never removed for so many reasons (including GPs KPIs, things never being removed from ntoes, not good descriptions of PEM meaning patients didn't have the lagnuage to explain symtpoms and felt desperate etc).

This is the risk with any research, sadly. There will always be some patients who are wrongly included in some trials, and some who get wrongly excluded. We don't know what we don't know, right?

I see why there is a list of exclusions, though. Whether or not we agree with it is another matter.

The hope is that looking at that "purer" signal would help us isolate what ME is. If we can do that, then excluding based on other conditions may not matter as much, as we would be able to determine if someone has ME regardless of what else they have.

This is all hypothetical, of course, because the fact of the matter is that exclusionary conditions and even diagnosis currently vary wildly across trials. There's very little consistency in what people look at.

 

[bobbler]

i am confused.

it probably all makes sense though. prob just me.


in the state of texas, there is a rumored needed killin' law. if there is a paywall, it needed removin'.


seid or something said somehting like it is a positive diagnosis, not a diag of exclusion. contradiciton.

the word comorbid has 16,000 definitions. without a definition, it is hard to interpret. m.e. causes disease a. or helps cause it. or a is unrelated. or disease a causes or triggers m.e. etc. which is it? whhy is it exculsionary?

m.e. is a @!#$ SERIOUS disease. it is multisystem, complex, and systemic. it attacks practically everything. yet the idea is that there is a coherence to it. ok.... let's buy that for a minute.

where do we get off knowing it isn't m.e. just because you have disease a ALSO? are researchers studyig severe? ok maybe we aren't knowing it. we are just positing it for hte sake of a research study. i can get behind that. except the other stuff i am saying.


otoh, if we present as classic b12 symptoms, if ther eis such a thing as classic, and it HAPPENS to overlap an m.e. set of criteria, does it?, then we can say after working up for b12 that it is exclusionary? is that the idea?


you say exclusionary disease a can cause all the symptoms described by the pwme in the 2.6 seconds allotted in the visit while the doctor is looking at the nurse's new uniform and thinking of the car he is going to buy and lunch.

while the pwme has oi and had to rest for a month to get to the appt and could have permanent harm. is the pwme going to list everything? or reember everything? couldn't that change the outcome?


where was i? m.e. can CAUSE all sorts of things. are we SURE it cannot also ause a bunch of other things? ok so you say those other things are a covering set for that pwithoutme so it isn't m.e. even so, could that be false? hickam vs. occam.

what about studying those whose m.e. has lasted for many many years and thus comorbidities by whatever definitsions? could the list chnge?

i'm sure there are pefectly respectable answers to my questions. he is talking about research definitions not just clinical. statndardiaztion can be good in principole. so yeah, let's eliminate outliers. just make sure 1] they are outliers, and 2] we circle around back to them.

i'd ust like in my brain not working state for this to make omre sense. i do not think this is obvious stuff. and when somthing is not obvious i think it needs careful wording, especially to get rid of ambiguities and be transparent about logic step by step if nec.

Agreed. LOok at Lupus, Sjogrens, other autoimmunes for example. Or other diseases. Arthritis has different types that go across from autoimmune and psirotic to osteo. There will be atypical types, different areas that get affected and 'related conditions' that generally given an indication you might also have something else.

I think trying to separate out those who have certain other things is a fools game. There isn't the precision until someone has looked and seen whether these comorbidities are 'new types' or ME specific types or common due to something underlying in both illnesses ie unusual not to develop one if you get the other.

Maybe back when it was 'fatigue' the exclusions list could be justified. But even anemia doesn't give you PEM so that list and train of thought is no longer justified in the same way. Currently this, if it is to exlcude from research or treatment, rather than for the good of the patient of just checking for other things that can be treated seems a big and not useful distraction.

Documenting these connections and being keen to get them right in a catalogue of common comorbidities by correcting those things that turned out to be misdiagnoses for ME in the old days, so if depression was an old wrong diagnosis it gets removed as 'commonly associated with' by people being re-checked and that being struck out, is a useful part of research surely to find out what it is.

But also because if ME comes with certain other conditions to have advice to check for them as time goes on (I know quite a lot who ended up with RA for example in the end).

The crtieria, properly defined (and then plotted) is really pretty distinctive. Yes we might find there are a few more unique conditions within it - but I have no issue with research managing to join forces in finding these within PEM and picking apart how these altogether result in similar effects, because they all somehow must be connected in leading to the same distinctive issues in some way.

And if we do end up managing to just be really strict on the presence of proper PEM/PESE then we have a really good start on finding out what on earth this is and who it covers and what else it tends to cause.

And also how bad the medical system has been - this list might as well be one of 'common misdiagnoses', that is there to ensure people are checked for these in case they do have them and these could be helped. Noone who this does mean a fix for is going to still go around claiming they have ME if all that goes away.

 

[bobbler]

This is the risk with any research, sadly. There will always be some patients who are wrongly included in some trials, and some who get wrongly excluded. We don't know what we don't know, right?

I see why there is a list of exclusions, though. Whether or not we agree with it is another matter. Don't get confused thinking that good science and research is all about exclusions. Unless you know for sure what % of those with ME have been investigated all their lifetime and never got another condition, and can be sure they are the rule rather than the exception? It feels a big old distraction to me.

Particularly when you have, for those with the diagnosis, the ability to check if it was a misdiagnosis anyway, and to have notes alongside so it doesn't sully data at all.

I get what you are saying re: hope is that looking at that "purer" signal would help us isolate what ME is, but the fallacy is that is a 'purer' type of ME for not having anything else, which isn't correct.

If we can do that, then excluding based on other conditions may not matter as much, as we would be able to determine if someone has ME regardless of what else they have.

This is all hypothetical, of course, because the fact of the matter is that exclusionary conditions and even diagnosis currently vary wildly across trials. There's very little consistency in what people look at.

NO I think research can be straightforward. We know ME/CFS is PEM/PESE and we should be able to do very good criteria for this.

If the research itself is for example going to be testing people walking upstairs then there might be justification for excluding those with conditions where the reaction to this or the treatment might be very different e.g. arthritis of the knees.

But I do not think it is methodologically correct to start cutting out genetic types or types or huge chunks of what ME is in the risk of actually getting a niche of people who don't represent the condition as it tends to be at all.

I think only going with 'pure' is as likely to end up with a misleading description/research - samples are supposed to represent a population. ME isn't a disease of 'not having anything else' as far as I know. If you do not have the stat on whether the 'pures' are indeed pure and are the biggest % of those with the illness then the idea they are the closest and most useful is wrong. Plus you don't know enough about who the 'pures' even are to know whether they will stay that way (ie you are only looking at early stage) and aren't a niche with some genetic uniqueness vs 90% of those with ME.

Lots of autoimmune conditions have common comorbidities, to think that you'd separate out those with sjogrens from lupus or RA research without fully knowing those conditions seems short-sighted and misleading.

Particularly when those with said diagnoses make it easy because if the bits that might be due to those illness appear as different peaks the ones who come with the diagnosis 'explain it' - and it certainly doesn't mean you wouldn't have 200 other people also appearing on that peak but just not with the diagnosis yet. You might then find those with said diagnosis had ME 5yrs longer than the others - hey presto a connection to look into etc. So they really do offer massive amounts of information in what the condition is.

 

[Trish]

NO I think research can be straightforward. We know ME/CFS is PEM/PESE and we should be able to do very good criteria for this.

If the research itself is for example going to be testing people walking upstairs then there might be justification for excluding those with conditions where the reaction to this or the treatment might be very different e.g. arthritis of the knees.

But I do not think it is methodologically correct to start cutting out genetic types or types or huge chunks of what ME is in the risk of actually getting a niche of people who don't represent the condition as it tends to be at all.

I think that depends what type of research is being done. If, for example, a research project is studying immune cells in people with ME/CFS, it's important to exclude from that study anyone who has a known immunological condition, so the abnormalities in their immune cells due to that other condition aren't confusing the data for ME/CFS.

They may have PEM and be accurately diagnosed with ME/CFS as well as their immunological condition, but including them makes it harder to tease out from the data what is due to which condition. There usually isn't enough funding to allow for the bigger sample sizes that would be needed if the study had to include everyone with ME/CFS who also had other conditions that are likely to affect the results and to study them as separate subgroups. So the only way to ensure the 'cleanest' sample possible is to exclude from the study those known to have the overlapping conditions.

Of course there may be a few wrongly included who have other conditions, but researchers need to try to get as clearly identifiable samples as possible.

 

[adambeyoncelowe]

Trish is right that it requires common sense. Most criteria are supposed to have a little bit of leeway involved for clinical judgement, but this should also apply to exclusionary criteria.

There would be times arthritis may not be an exclusionary diagnosis for a bit of research and times when it absolutely would. Mobility might be a time to exclude; breathing issues, for example, might not, but then you might exclude asthma in that second group as well.

I would also be wary of trials that selected patients based only on PEM, until we've tested the supposition that PEM only occurs in ME.

It would make more sense to diagnose patients according to current international criteria and either subgroup the results based on different criteria used (so an IOM group, a CCC group and an ICC group, for example) or get agreement about which criteria researchers should use, to keep things consistent, with a view to adjusting things in future if we need a broader or narrower perspective.

 

[Nightsong]

This paper, and specifically the table below, was recently referenced in a so-called "consensus statement" about a research definition for ME/CFS (link to first post); the first author is also one of the authors of the statement and two of the other authors were among the initial signers. The list of agreed-upon exclusionary conditions is given as follows:



I've already formulated an opinion but would be interested to know what others make of it.

 

[duncan]

I see under infectious diseases they've listed Lyme disease and parasitic diseases. How do they definitively know whether someone actually has Lyme disease or a babesiosis strain or a bartonella strain?

Seems like they are comfortable building in a number of assumptions that may muddy the waters, and potentially undo their bid to strengthen the rigor of ME/CFS research.

You cannot exclude a disease, any disease, that produces ME/CFS-like symptoms unless you can prove that disease is not there.