I think false positive PEM identification from GPs is an extreme rarity. And I do not mean that yo can give an ME/CFS diagnosis based on PEM only. Of course you need the rest of the symptoms that is required -whatever criteria is being used. My main point is that as of now I think M;E/CFS is severely underdiagnosed. Furthermore usually extremely delayed if given. As of now I see no sign of hospitals/ consultants/ specialists moving towards acknowledging ME/CFS or tacking responsibility for the disease. So that leaves me with focus on GPs.
Should we initiate development of a new, short questionnaire to identify PEM (to aid diagnosis)?
Jonathan Edwards
Senior Member (Voting Rights)
There was a study from the USA which, if I remember rightly, indicated that doctors overdiagnosed ME/CFS quite a lot, in that most people with the diagnosis on their records did not think they had it. I suspect there is both marked underdiagnosis and overdiagnosis, depending on the doctor.
Utsikt
Senior Member (Voting Rights)
I think we’ve seen several studies showing that self-report of ME/CFS is much more prevalent than self-report a doctor’s diagnosis of ME/CFS, which is much more prevalent than fulfilling e.g. the CCC criteria based on questionnaires. This indicates that over-diagnosis is prevalent.
But I have no doubt that there are also people that meet the criteria that have not received the diagnosis, or that have received an FND, MUPS, BDD, PPS, etc diagnosis instead. Or just depression, anxiety, etc. The average time to an ME/CFS diagnosis is >5 years in Norway if I remember correctly.
Both situations need solution, but ideally the solution for one should not make the other much worse. Which is why I’m apprehensive about questionnaires or sets of questions without also making sure the false positive rates aren’t too high.
I agree that it might be viable to focus on GPs in the short term in Norway. I fear we’ll need an effective treatment to get the psychosomatics to back off, or at the very least to get someone else to «claim» us. The authorities have shown they are incapable of seeing through their charades, or that they simply don’t care.
The lack of intense rants here about it being worse?
I think that's a possible factor in determining PEM. I can't recall seeing a poll about abruptness of PEM onset. For me, PEM was quite abrupt: baseline ME symptoms one minute, then feeling them flare up dramatically. I'm guessing that there are PWME with more gradual increases, which leads to whether there are multiple mechanisms that lead to an increase in symptoms following a trigger.
Depending on the study's goal, there should probably be study-specific criteria for PEM. While that makes studies more difficult to compare, we don't have a good enough understanding of PEM for that.
I see that as pre-defining PEM based on an assumption (that there's an energy limit). My PEM did not seem to be based on energy limit; it seemed to be based on exceeding the accustomed strain on muscle tissue. A minute of unaccustomed but not very energy intensive exertion would trigger my PEM, while hours of accustomed highly energy intensive exertion wouldn't. While many PWME may correlate triggering with energy (ATP) usage, that might be a false correlation. It might correlate with muscle microtearing (and subsequent immune response) or gut microbiome alteration or vagus nerve activity or whatever else that occurs from activities.
Nightsong
Senior Member (Voting Rights)
In relation to diagnostic unreliability I recall two other salient papers: the first being the Samms-Ponting paper assessing various different sub-cohorts from the UK Biobank and their questionnaire responses which showed high phenotypic instability.
The second was the Johnston et al. paper from 2016 which found that, of "535 patients diagnosed with CFS/ME by a primary care physician" in an Australian cohort, "30.28% met Fukuda criteria. A further 31.96% met both Fukuda criteria and International Consensus Criteria. There were 14.58% reporting chronic fatigue but did not meet criteria for CFS/ME and 23.18% were considered noncases due to exclusionary conditions".
Of course, what is the goal is both lack of under-diagnosis and over-diagnosis of ME/CFS. I think a reasonable goal is correctly identifying and diagnosing ME/CFS based on CCC criteria in adults and Jason 2006 criteria in children and adolescents. In Norway I am convinced that there is mainly a definite under diagnosis of both and little over-diagnosis. I also think that applies to most (all) of Europe. For an acceptable situation you need a definite improvement at the GP level, hospital/ consultant level and in the general population about the reality of this disease. As part of a long term strategy to improve that situation I think improving understanding and identifying PEM is paramount. So I do not think it is that important to estimate how much/ how often a person had PEM. I think the first target is identifying it in a yes/ no manner. There is a definite limit on capacity for activity for every pwME/CFS. If activity level (either due to one activity or the sum of activities) exceeds that limit it results in PEM.
So a search for good content in questions a doctor could pose to uncover PEM to a patient is very important. Typically the background would be a new onset (even though that could be years ago) of definite reduction of functional capacity and persistent fatigue. Even though identification of the other ME/CFS symptoms are extremely important a lack of PEM (assuming qualified and correct identification) excludes ME/CFS. I think pwME/CFS (given correct diagnosis) and clinicians with extensive experience with ME/CFS are the main sources for this search. I am then excluding the experience of clinicians with a BPS view of the disease as a source.
I definitely agree that in. very severe ME/CFS PEM may be unidentifiable due to the extremely reduced capacity for all activity and the wise avoidance of triggering PEM. Still I have never, in such patients, found a problem with identifying definite in the earlier course of their disease.
Research on all aspects of PEM is, however, I think extremely important including biological aspects/ mechanisms, clinical subtypes, importance PEM frequency/ severity for clinical course, and more. Still, you first have to identify PEM presence reliably.
Thanks for all input!!
So a search for good content in questions a doctor could pose to uncover PEM to a patient is very important. Typically the background would be a new onset (even though that could be years ago) of definite reduction of functional capacity and persistent fatigue. Even though identification of the other ME/CFS symptoms are extremely important a lack of PEM (assuming qualified and correct identification) excludes ME/CFS. I think pwME/CFS (given correct diagnosis) and clinicians with extensive experience with ME/CFS are the main sources for this search. I am then excluding the experience of clinicians with a BPS view of the disease as a source.
I definitely agree that in. very severe ME/CFS PEM may be unidentifiable due to the extremely reduced capacity for all activity and the wise avoidance of triggering PEM. Still I have never, in such patients, found a problem with identifying definite in the earlier course of their disease.
Research on all aspects of PEM is, however, I think extremely important including biological aspects/ mechanisms, clinical subtypes, importance PEM frequency/ severity for clinical course, and more. Still, you first have to identify PEM presence reliably.
Thanks for all input!!
Utsikt
Senior Member (Voting Rights)
I really appreciate that you engage here, your work in Norway has been invaluable and some of the short videos have been crucial to get my carers to understand certain aspects of living with ME/CFS.
I think people looked into the estimated prevalence of ME/CFS when making the factsheets, and a rough estimate was that it was 1/250 = 0.4 %. If we say 0.3 - 0.5 % to add a reasonable margin because these numbers can’t be exact, the conservative estimate based on only non-GP diagnoses is still much higher than expected. We might have as many as 50 % false positives when adding the GPs if the numbers are representative.
The remarkable thing is that barely any of the G93.3 cases of Kielland returned to their pre-illness income level, which means that they are still sick with something chronic or untreated. If a substantial share of them are false positives, a not insignificant amount of people might be sick for no reason other than being given the wrong diagnosis.
I understand that you see a lot of the false negatives in you practice, so it’s fair that you might focus on those. I can only look at the overall numbers.
Things might have changed since 2018. There was a recent paper that showed that psychiatrists can’t agree on which diagnoses to use in most cases, and I would not be surprised if that’s also the case for ME/CFS. I suspect some units use it a lot more than others, and the BPS folks are pushing for using FND/PPS/BDD. Owe has been denying ME/CFS diagnoses for a long time at Haukeland, maybe you end up with a lot of then afterwards?
Kielland found that between 2016 and 2018 an average or 1850 received a G93.3 diagnosis per year, which excludes GPs. If we multiply that by 20 to get a conservative estimate of the total prevalence of ME/CFS, we get 37 000 or 0.7 % based on a population of 5.3M in 2018.
I think people looked into the estimated prevalence of ME/CFS when making the factsheets, and a rough estimate was that it was 1/250 = 0.4 %. If we say 0.3 - 0.5 % to add a reasonable margin because these numbers can’t be exact, the conservative estimate based on only non-GP diagnoses is still much higher than expected. We might have as many as 50 % false positives when adding the GPs if the numbers are representative.
The remarkable thing is that barely any of the G93.3 cases of Kielland returned to their pre-illness income level, which means that they are still sick with something chronic or untreated. If a substantial share of them are false positives, a not insignificant amount of people might be sick for no reason other than being given the wrong diagnosis.
I understand that you see a lot of the false negatives in you practice, so it’s fair that you might focus on those. I can only look at the overall numbers.
Things might have changed since 2018. There was a recent paper that showed that psychiatrists can’t agree on which diagnoses to use in most cases, and I would not be surprised if that’s also the case for ME/CFS. I suspect some units use it a lot more than others, and the BPS folks are pushing for using FND/PPS/BDD. Owe has been denying ME/CFS diagnoses for a long time at Haukeland, maybe you end up with a lot of then afterwards?
I don't think a rough estimate of 0.4% prevalence of ME/CFS is likely, even before the pandemic. I think it is higher. I am afraid, as of now, it may be above 1% when post covid data included.
All that (prevalence) is still not really of interest for my point. Under- or over-diagnosis are both highly undesirable. So my interest is how to improve identification of PEM by GPs, or any clinician/ doctor for that matter. The question contents during a consultation is extremely important. Improvement of that is what I am targeting.
My suggestion for content of the opening question is this (as described previously:
"If you do very much more than you usually do on a given day because you had to or wanted to, what happened the day and days afterwords? Do you feel better and can do a bit more of activities? The same as before? Worse, with less capacity for activities?". I would then ask specifically for motor activities and social/cognitive activities or both.
Jonathan Edwards
Senior Member (Voting Rights)
My guess is that everyone, including normal healthy people, will say they are worse after, maybe for a few days and maybe not hitting one ntil a while after activity.
That's why it's just an opening question. The doctor would go on to find out their usual activity levels and the symptoms of worsening after activity and how it affects function. If they listen carefully it should be easy to distinguish between PEF/DOMS in a healthy person, and the symptom and activity pattern in ME/CFS.
Jonathan Edwards
Senior Member (Voting Rights)
OK, that's fair enough.
I just worry that couching the questions in terms of post-activity may not be ideal. I realise that is how PEM is defined but that then begs the question of whether the way PEM is defined is ideal.
I think the point of 'abruptness' is interesting. I have a sense that the key feature is suddenly feeling bad, more than any particular relation to activity. I don't think I have had what people call PEM with my Longish Covid. I would have said I was worse, more exhausted, but I did not have the sense of badness hitting me, just being knackered. In contrast, after EBV I clearly remember abruptly feeling bad after a minor sports outing (more than usual activity but not under any pressure or enough to be tiring normally).
The sense was "Hey, what hit me, why am I still unwell". I am not sure that I could have gauged a loss of function. I just felt that no way did I want to go out canoeing again if I was going to feel like this after. Maybe that is not ME/CFS PEM either, but if not I wonder whether we are assuming that there is a specific phenomenon when things are much more variable over a spectrum.
Kitty
Senior Member (Voting Rights)
Before I encountered the concept of PEM I talked about cycles.
How ill I feel at any moment has always varied hugely, and it seemed as if there were at least four cycles of fluctuation in operation: across single days, groups of days, groups of weeks, and periods of about two years. The shorter cycles probably reflect what we describe as PEM, the longer ones are more about underlying severity.
Is it possible the cycling is what's distinctive, rather than what triggers it or how it makes us feel?
How ill I feel at any moment has always varied hugely, and it seemed as if there were at least four cycles of fluctuation in operation: across single days, groups of days, groups of weeks, and periods of about two years. The shorter cycles probably reflect what we describe as PEM, the longer ones are more about underlying severity.
Is it possible the cycling is what's distinctive, rather than what triggers it or how it makes us feel?
Peter T
Senior Member (Voting Rights)
Intuitively this cycle idea also makes sense to me, though I would add a longer cycle of six to ten years. I think in terms of PEM and the severity of the underlying ME/CFS.
Utsikt
Senior Member (Voting Rights)
Kielland’s data was pre-pandemic, so that means there are probably more false positive pre-pandemic cases in Norway than true positive cases.
I agree that the rate has increased following covid, but 2.5x seems like a lot. If you count the transient cases that recover within a couple of years it might be more realistic.
We already have questions that catch almost all PEM cases (effectively eliminating underdiagnosis), like the DSQ, so shouldn’t our focus be on reducing the false positive cases?
I don’t understand how that can be achieved without first educating the GPs about what ME/CFS looks like, and what it probably does not look like. Otherwise you’d end up with a lot of false positives when using questions, and that will further harm the people with ME/CFS because then GPs will think they have experience with ME/CFS when they are actually looking at something else.
Kitty
Senior Member (Voting Rights)
I suspect GPs would resist questionnaires anyway. They know diagnosis is more of a process than an event, and it can take months or years. It tends to involve more than one medical professional, initial clinical suspicions are often wrong, and important information may only come to light after several interactions. If diagnostic questionnaires really worked we wouldn't need nearly so many doctors.
In any case, there's an obvious false positive (self resolving post-viral illness) that probably can't be excluded except by the old fashioned watch-and-wait method. If we want ME/CFS diagnosed early to avoid possible exacerbation from pushing through, we're likely to catch a proportion of those cases initially.
Utsikt
Senior Member (Voting Rights)
The approach to PVF is the same as to ME/CFS so that’s probably fine, as long as we give realistic prognoses (probably fairly decent some months in) while emphasising the uncertainty and the need to adapt now and not just endure crushing symptoms for a while and expect it to resolve.
Absolutely, but a list of questions doesn’t have to be a questionnaire. There are good ways to ask about something, and there are bad ways. Skipping past 90 % of the trial and error by learning from others is probably a good thing.