Preprint Epidemiology of Myalgic Encephalomyelitis among individuals with self-reported CFS in BC, Canada, and their health-related quality of life, 2024,Nacul

https://www.medrxiv.org/content/10.1101/2024.05.16.24307437v1


Epidemiology of Myalgic Encephalomyelitis among individuals with self-reported Chronic Fatigue Syndrome in British Columbia, Canada, and their health-related quality of life

Enkhzaya Chuluunbaatar-Lussier, Melody Tsai, Travis Boulter, Carola Munoz, Kathleen Kerr, Luis Nacul
doi: https://doi.org/10.1101/2024.05.16.24307437

Abstract

Background:
There is no accurate data on the epidemiology of Myalgic Encephalomyelitis/ Chronic Fatigue Syndrome (ME/CFS) in Canada. The aims of the study were to describe the epidemiology of confirmed ME/CFS cases and their health-related quality of life (HRQoL).

Methods:
This is a cross-sectional study with British Columbia Generations Project (BCGP) participants who self-reported having CFS and population-based controls with no fatiguing illness. Participants completed the Symptoms Assessment Questionnaire, RAND 36-item Health Survey, and Phenotyping Questionnaire Short-form. These assessments enabled the identification and characterization of confirmed cases of ME/CFS. Those with self-reported diagnoses who did not meet study diagnosis of ME/CFS were subcategorized as non-ME/CFS cases.

Results:

We included 187 participants, 45.5% (n=85) self-reported cases and 54.5% (n=102) controls; 34% (n=29) of those who self-reported ME/CFS fulfilled diagnostic criteria for ME/CFS. The population prevalence rates were 1.1% and 0.4% for self-reported and confirmed ME/CFS cases respectively. Participants displayed significantly lower scores in all eight SF-36 domains compared to the other groups. Mental component scores were similar between ME/CFS and non-ME/CFS groups. The main risk factor for low HRQoL scores was fatigue severity (β = -0.6, p<0.001 for physical health; β = -0.7, p<0.001 for mental health).

Conclusions:

The majority of self-reported cases do not meet diagnostic criteria for ME/CFS, suggesting that self-reported CFS may not be a reliable indicator for a true ME/CFS diagnosis. HRQoL indicators were consistently lower in ME/CFS and non-ME/CFS cases compared to controls, with ME/CFS cases having lower scores in most domains. Having higher symptom severity scores and perceived poorer health were the significant affecting factors of lower HRQoL. Although self-report can be used as screening to identify cases in populations, we suggest studies of ME/CFS should include appropriate medically confirmed clinical diagnosis for validity. Further large-scale population-based studies with simultaneous medical assessment are suggested to further characterize validity parameters of self-reported diagnosis.

 

Wondering how they found a sample of people who fit the CCC and/or IOM, of which >40% rated their health as “good”?

Also here is link to pdf if anyone wants: https://www.medrxiv.org/content/10.1101/2024.05.16.24307437v1.full.pdf

 

Ignoring the jumbled confusing use of ME, CFS and ME/CFS in the abstract.

"Consenting individuals completed the following three questionnaires: 1) Symptoms Assessment Questionnaire (SAQ) for diagnosis confirmation, which is a symptom specific questionnaire for ME/CFS diagnosis aid [25]; 2) Phenotyping Questionnaire Short Form (PQsym-12), a symptom severity assessment with 12-questions created by the CureME Group in the United Kingdom used to collect data enabling clinical phenotyping information, with higher scores representing greater severity of ME/CFS symptoms [25]; and 3) RAND 36-Item Short Form Health Survey (SF-36) to assess physical and mental function, with scores ranging between 0-100 for eight domains, and two summary normalized scores for overall physical and mental functions, with higher scores representing better health [26].

Case Definitions of ME/CFS

In the absence of biomarkers for diagnosis, case definitions of ME/CFS were based on symptoms presented by participants. Responses to the SAQ enabled the identification of “confirmed cases” of ME (categorized as the “ME/CFS group”), as it contains a built-in algorithm that enables categorization of participants according to whether they meet the diagnostic criteria of the 2003 Canadian Consensus Criteria (2003 CCC) [27] and/or 2015 Institute of Medicine (2015 IOM) criteria [28]. Those with a self-reported diagnosis, but who do not meet the study diagnosis of ME, were sub-categorized as the “non-ME/CFS group.”"

 

This seems to be the main conclusion:

"...34% (n=29) of those who self-reported ME/CFS fulfilled diagnostic criteria for ME/CFS. The population prevalence rates were 1.1% and 0.4% for self-reported and confirmed ME/CFS cases respectively

[...]​

"Our study showed that the use of self-reported diagnosis of ME/CFS is not a reliable indication to estimate the prevalence of this disease and is not well suited for clinical or epidemiological studies, due to the potential for bias as a result of misclassification of disease status" ​

 

Unfortunately, the follow-up examination of people who self-reported CFS did not include an clinical examination by a physician but more questionnaires.

Does anyone have acces to SAQ or PQsm-12?

 

Twitter thread summary of the preprint:
https://twitter.com/user/status/1792196610870710657


1) Interesting new study from Canada from the research team of Luis Nacul.
The authors recruited participants from ‘CanPath’: a population-based cohort that includes data from over thousands of Canadians.

2) In 2016 19,145 of these participants responded to the question about a Chronic Fatigue Syndrome diagnosis. Of those participants 1.1% responded ‘yes’ to ever having been diagnosed with CFS.

3) Nacul et al. did a follow-up study of the participants who reported a CFS diagnosis to let them fill in some additional questionnaires and check the accuracy of their ME/CFS diagnosis.

4) They found that only a third of participants with self-reported CFS, fulfilled ME/CFS diagnostic criteria (IOM and/or CCC).
The estimated prevalence therefore dropped from 1.1% to 0.4%.

5) The authors conclude:
“Our study showed that the use of self-reported diagnosis of ME/CFS is not a reliable indication to estimate the prevalence of this disease and is not well suited for clinical or epidemiological studies….”

6) There are lots of caveats about this study: it only checked ME/CFS diagnosis using questionnaires, not a clinical examination and the response rate was quite low (43%).

7) The prevalence estimate of 0.4% is also likely an underestimation as it does not take into account patients who meet ME/CFS criteria but are undiagnosed.

8) Nonetheless, interesting that we finally have a study that attempted to check the accuracy of self-reported CFS in population cohorts.
This may be relevant to the Lifelines population cohort in the Netherlands which plan to study ME/CFS as well.

9) The study has not been peer-review yet. The pre-print is available here:
https://www.medrxiv.org/content/10.1101/2024.05.16.24307437v1.full.pdf

 

So the options are "good" or "not good"

Initially I agreed with you (in my head). But then I recall answering a similar questionnaire and nearly putting my health down as good. While I'm not bedbound, I can only spend a small amount of time sitting upright per day and need a mobility aid to go over 25 m or so. And of course I have lots of symptoms and impairments. So it's a bit ridiculous for me to think that my health is good.

But you can get used to this and think more in terms of your general health.

And most people would be less severely affected than me.

So I'm not sure if this gives us definitive evidence.

 

Yes. In some twisted way I feel my health is good, despite the ever present, highly unpleasant - understatement - and very limiting ME symptoms. I think it's because I've had ME so long it's just my normal. So when asked about my health I think about all the other common illnesses I don't have. No arthritis, no heart disease, no diabetes, no cancer, etc etc etc

Goes to show how careful one has to be when interpreting questionnaire answers

 

That seems to me a massive missed opportunity. The dark figure of the undiagnosed is arguably an even bigger source of uncertainty than those diagnosed without meeting the criteria.
There must be a way, in a cohort like this one, of identifying potential missed diagnoses in the existing questionnaire data and then pull them out for further investigation.

 

The Jason et al. estimate of 0.4% was for the Fukuda CFS criteria so it seems that self-reported CFS is much higher than actually meeting CFS criteria. It is still a bit unclear to me how this is possible, given that most of the 0.4% were undiagnosed.

It would be great if a study could help clarify this: what's the situation with all the people saying they had a CFS diagnosis but not meeting the criteria: did they initially meet the criteria but then improve enough to no longer meet it? Or is something else going on. It might also has something to do with the vague term 'chronic fatigue syndrome' being given to various types of unexplained fatigue that it was not intended for.

 

It's great to have an assessment on what proportion of people who report a "Chronic Fatigue Syndrome diagnosis" meet ME/CFS criteria. 34% is unsurprising. People might remember a "chroninc fatigue" diagnosis as CFS, bu we also know that around half of GP diagnoses are wrong.

I don't think we can safely conculde much about prevalence from this study - because the British Columbia Generations Project itself has a heavily skewed demographic profile: 69% are female as well as being better educated, more likely to be white and with higher incomes than the age matched BC population. It also has higher reported levels of chronic illness than age matched population controls. And only includes people aged 35-69. This isn't a suitable cohort from which to assess prevalence.

The good/bad health answers threw me - and I would never have answered a health survey describing myself in good health, but I have to agree with the comments above. Not least because of the small confidence intervals on the phenotyping questionnaire and SF36 scores. If some people really had good health, that would be reflected in much wider intervals.

 

Do you have any further references for this?

I remember a couple of studies showing that approximately half of cases sent to ME/CFS tertiary centres did not fulfil ME/CFS criteria. But I don't think the GP's diagnosed these patients with CFS, it were patients suspected of having ME/CFS.

But perhaps you were referring to something else.

 

From a recent conversation with my GP's clinic, it appears that the diagnosis on my medical record is fibromyalgia. Probably. When I talked with an assessor for disability years ago, they mentioned chronic fatigue. I don't know what's officially coded with the system, it could be neither. Hell, it could be depression for all I know.

Health records are a tainted source of information here, they must be considered unreliable. As in literally they should not be used at all, not even to compare with, other for the % of people who do get diagnosed, which is a property of what clinicians do, not what affects patients. Which is a huge problem in itself, the number of things that must go wrong for this to happen is simply absurd, this is a completely unique problem in the professional world, but this is the reality we live with.

Not that this is especially surprising given the last few decades. It turns out that lying and making stuff up is not a good way to do health care. Who knew? But then you look at what's been happening with the pandemic, especially the huge number of discussions lately over children missing school and worker shortages, with ZERO mention of COVID, even with many studies plainly showing that it's illness-related and research showing all the problems COVID and other infectious diseases cause, you sort of get the idea that this is basically common in health care, that medical data are pretty much unreliable as a default. Which is astounding, but this is the reality we live in.

 

References from my post:
half of all the referred patients to a specialist CFS clinic had alternative medical and psychiatric diagnoses (2); Newton 2010 found similar.
One of those was for suspected cases, I'm pretty sure the other was of diagnoses but also see ref that found 48% of GPs were not confident making a diagnosis. Also, data from the CPRD (Collin, 2017 ) shows a sex ratio of 2.3 for recorded diagnoses in primary care, which is suspect to say the least.

 

Given these numbers, I wonder how accurate that 3.3 million figure that CDC published is.

Although they asked patients if they “still have ME/CFS” in the CDC questionnaire, am unsure if this one counted only those who had been diagnosed and not those who said they still had.

 

The M.E specialist I saw in 1991 based his assessment on onset, history, r/o other causes and relapses. The diagnosis was not made based on symptoms.

 

You know, whenever I get that "How would you rate your health?" question it is hard for me to say "poor" even though I'm disabled.

Part of it is denial and internalized ableism, not to mention society's expectation that I should be positive and upbeat.

Part of it is knowing so many other folks (a few in person, many more online) who are so much sicker than I am.

Part of it is that some of my issues have been solved over the years. I used to have intense endometriosis pain - thank goodness excision surgery removed almost all the pain.

Part of it is that I remember being *so much worse* in 1991. I couldn't walk across the room.

I always want to say, "Good/fair/poor? Relative to what?"

Yeah, I know. I always over-analyze these things.