Rates of ME/CFS following Covid-19

[Adrian]

Thread split from [UK] A proposal for an ME/CFS, Long Covid, and Post-Infectious Disease research platform
about a proposal that mentioned that around 50% of people with Long Covid have ME/CFS.

I think it’s one of those memes that people have seen repeated so many times that they just become an agreed upon truth no matter what the data says.

Like “25% of people with ME are housebound or bedbound.”

There have been a number of studies Jason did one Estimating ME/CFS prevalence in individuals with long COVID. I think there was one study which had smaller numbers.

They could have done without the last part.

The point about including long covid is there are probably common mechanisms (given ME criteria are met for a significant number of people) and hence it makes sense to develop research capacity jointly. It also extends the numbers and the size and justification of the problem

 

[Utsikt]

The point about including long covid is there are probably common mechanisms (given ME criteria are met for a significant number of people) and hence it makes sense to develop research capacity jointly. It also extends the numbers and the size and justification of the problem

I understand why they want to mention LC, but saying that 50 % with LC also have ME/CFS is simply wrong. Those numbers are from reports from specialist clinics and they can’t be generalised. There really is no excuse for mistakes like that.

It would have been more then enough to say that the number of people with ME/CFS is increasing due to frequent covid infections.

 

[Utsikt]

@Adrian regarding the study on prevalence of ME/CFS in LC patients: the study that found 58 % recruited patients from SoMe LC groups. That’s a massive source or bias.

https://www.s4me.info/threads/me-cf...ients-with-long-covid-2022-jason-et-al.31036/

 

[Adrian]

I understand why they want to mention LC, but saying that 50 % with LC also has ME/CFS is simply wrong. Those numbers are from reports from specialist clinics and they can’t be generalised. There really is no excuse for mistakes like that.

It would have been more then enough to say that the number of people with ME/CFS is increasing due to frequent covid infections.

The numbers are uncertain and my guess would be higher. The 50% does come from what few studies there are. The 400k for ME comes from Chris Ponting's latest paper (I think published today). But for years we have talked about a vague 250k in the UK. The important thing is numbers give an idea of the order of magnitude of the problem and hence the need for solutions.

And the latest Jason figures are from patients recruited from social media

Considering the limitations of previous research, Dr Jason and Dorri aimed to make improvements in the methodology in order to determine the proportion of individuals with long COVID meeting “the ME/CFS case definition”, including PEM. The study recruited 465 participants, predominantly from social media sites dedicated to COVID-19 and communities and groups related to long COVID. The participants completed three types of DePaul Symptom Questionnaire (DSQ): DSQ-COVID – a new questionnaire measuring common symptoms of COVID and long COVID; DSQ-SF – a short form (SF) questionnaire assessing ME/CFS symptoms relating to both Canadian Consensus Criteria and Fukuda criteria; and DSQ-PEM – a brief questionnaire assessing PEM.

Given lack of investment in ME/LC research then figures will always be difficult.

 

[Adrian]

@Adrian regarding the study on prevalence of ME/CFS in LC patients: the study that found 58 % recruited patients from SoMe LC groups. That’s a massive source or bias.

https://www.s4me.info/threads/me-cf...ients-with-long-covid-2022-jason-et-al.31036/

But similar results to recruitment through clinics.

 

[Utsikt]

The 50% does come from what few studies there are.

But those studies are not generalisable, and we can’t pretend they are just because it fits the narrative of «ME/CFS is a massive societal issue and it’s profitable for society to invest in research».

The important thing is numbers give an idea of the order of magnitude of the problem and hence the need for solutions.

And the latest Jason figures are from patients recruited from social media

We don’t need a bigger number than 250-400k to get the point across. The annual cost is already in the billions. It’s going to be profitable to solve the problem regardless of how many with LC that also has ME/CFS.

Given lack of investment in ME/LC research then figures will always be difficult.

Yes, but that doesn’t mean that we should use numbers that clearly are not appropriate for that use case. It’s like begging to not be taken seriously.

But similar results to recruitment through clinics.

And those numbers are also very, very biased and clearly not generalisable.

 

[Utsikt]

I'm surprised to see such scepticism on the 50% figure for the number of PwLC who fit an ME diagnosis, given that we were all warning about a mass-ME event when the pandemic started. But if that figure is anything like true,

Studies on self-selecting participants or patients at specialist clinics can’t be used to estimate the prevalence of anything. I don’t understand why that is surprising.

 

[Jonathan Edwards]

I haven't seen any reason to think that post-Covid ME isn't just plain old ME, but am I missing something?

The point is that post-Covid ME/CFS is probably a tiny proportion of Long Covid. Most of it is probably like mine - post-viral fatigue of a resolving sort.

 

[Sasha]

Studies on self-selecting participants or patients at specialist clinics can’t be used to estimate the prevalence of anything. I don’t understand why that is surprising.

I was lazily relying on this meta-analysis, but are the studies all poor-quality?

If they are, do we have any better studies that would suggest that a low proportion of PwLC have ME?

[Edit: Dubbo came up with 11% or so after other viruses such as EBV.]

 

[Sasha]

The point is that post-Covid ME/CFS is probably a tiny proportion of Long Covid. Most of it is probably like mine - post-viral fatigue of a resolving sort.

But don't the studies that look at it look at PEM as a criterion (which you said you didn't have, I think)?

 

[Utsikt]

I was lazily relying on this meta-analysis, but are the studies all poor-quality?

I have not checked the quality of the studies, but based on the descriptions they are not suitable for determining the prevalence of ME/CFS in LC patients.

And it received some pretty harsh critique here and elsewhere:
https://www.s4me.info/threads/the-p...ew-and-meta-analysis-2024-dehlia-et-al.40426/

do we have any better studies that would suggest that a low proportion of PwLC have ME?

We have studies suggesting that most pwLC recover within a year or two. And we can’t just assume that it’s high because it hasn’t been proven to be low (not saying you say that, just pointing it out).

 

[Jonathan Edwards]

I was lazily relying on this meta-analysis, but are the studies all poor-quality?

I think the analysis is probably useless. It looks at the proportion of people with LC who satisfy ME/CFS criteria. But the majority of people who get LC get better by six months or at least a year. If you sample the ones who go on being ill then quite a few may have ME/CFS. If the figure was really 50% it would show up on all sorts of statistics on absence from work etc. There has been an increase in people off sick but not to that extent (assuming that 10% of people who have Covid get a period of 'LC'.)

 

[Sasha]

Where have I expressed scepticism of that figure? What I'm pointing out is that there is a lot in the LC phenotype that isn't ME/CFS, and researching that could easily takeover this proposal.

Sorry, I gave the impression that I was addressing you when I was talking about the 50% but you had indeed not expressed any scepticism about it and I was addressing the scepticism elsewhere on the thread in order to lead up to my point about how bringing in LC could bring wider support.

I understand your concerns about the non-ME version of LC taking over but we're starting from a point of having almost nothing to lose. I see such barriers of stupidity to our emerging exciting researchers getting any money.

 

[Sasha]

I have not checked the quality of the studies, but based on the descriptions they are not suitable for determining the prevalence of ME/CFS in LC patients.

And it received some pretty harsh critique here and elsewhere:
https://www.s4me.info/threads/the-p...ew-and-meta-analysis-2024-dehlia-et-al.40426/

Thanks, I'll have a look. I may be being too casual about this!

We have studies suggesting that most pwLC recover within a year or two. And we can’t just assume that it’s high because it hasn’t been proven to be low (not saying you say that, just pointing it out).

I wonder if most PwME recover within a year or two. We can't, of course, assume it's high because we can't prove otherwise but this whole thing is really underlining what a basket case our figures are in general.

 

[Utsikt]

this whole thing is really underlining what a basket case our figures are in general.

Yup!

 

[Simon M]

do we have any better studies that would suggest that a low proportion of PwLC have ME?

We do. The 50% figure comes from biased samples, either specialist clinics or social media groups, who are likely to be more severe. And the massive numbers come when you apply the 50% figure to the ONS data.
It is the ONS data itself that undermines the 50% claim. Because the ONS studied also asked people if their symptoms limited their daily activities not at all, a little, or a lot (which is the SF 36 approach). Most people say their long Covid limited Their daily
activities either not at all or only a limited amount and that means they would not meet any definition of me/CFS the time I’m aware of.

You can make a case that the 50% figure applies to those who are “limited a lot” – but that’s a much smaller proportion of long Covid With me/CFS, and much smaller numbers overall.

I’m sure that Action for ME know this, I’m not sure why People are pushing a number that Doesn’t add up, other than it is a big number.

 

[Jonathan Edwards]

Do we know that most people who get non-Covid ME don't get better by a year or so, though, and aren't just like ME-type PwLC in that regard?

Yes I think we do know that. In fact I would say that we are only really interested in the people who have an illness that lasts more than a year and we can call that ME/CFS. The others are better described as having post-viral fatigue. The two illnesses may have major commonality in terms of physiology but it is the long term persistence of ME/CFS that makes it a major disabling illness and something different from PVFS.

 

[Sasha]

We do. The 50% figure comes from biased samples, either specialist clinics or social media groups, who are likely to be more severe.

I agree that those samples would be the more severely ill but is there any reason why a different proportion would have ME-type LC than at lower levels of severity? (Just thinking aloud about whether there is any way to rescue some useful information from these studies.)

Because the ONS studied also asked people if their symptoms limited their daily activities not at all, a little, or a lot (which is the SF 36 approach). Most people say their long Covid limited Their daily activities either not at all or only a limited amount and that means they would not meet any definition of me/CFS the time I’m aware of.

I don't think that the arbitrary ME diagnostic cut-off means that those who fall below it don't have ME, though.

 

[Robert 1973]

I’m sure that Action for ME know this, I’m not sure why People are pushing a number that Doesn’t add up, other than it is a big number.

People campaigning for anything have a tendency to believe and/or use unreliable data if it supports their objectives and vice-versa. I have always considered that to be unfortunate and wrong but there are numerous examples (not least in US politics) where it has be shown to be effective, at least in the short term. Apart from the value of truth per se, I consider it to be mistake to use invalid data to support valid arguments as it enables adversaries to undermine the latter by exposing the former – as has happened repeatedly with criticism of BPS research.

 

[Simon M]

but is there any reason why a different proportion would have ME-type LC than at lower levels of severity?

Yes. We have no data on whether or not these people have the same illness – no evidence at all. What we do know is that long Covid is “defined place great by a couple of hundred symptoms – it’s not based on a limited number of course symptoms. Some definitions of long Covid, which have been used to diagnose patients, don’t even require Significant physical disability. Mental health problems alone will suffice.

So long Covid is an incredibly broad thing, something the field doesn’t seem to have got to grips with yet.
there’s some evidence of studies that a lot of people who didn’t get Covid have long Covid symptoms seen afterwards (there’s an employment one and at least one other).

one of the major challenges with ME/CFS is identifying people that have something distinct, as opposed to people have general health, including chronic fatigue. Without any data, we can’t receive anything about this huge group of people with a broad range of symptoms.

I don't think that the arbitrary ME diagnostic cut-off means that those who fall below it don't have ME, though.

See above.

and it: there’s wide agreement that postexertional malaise is a defining symptom of the illness Because it’s so unusual.

It’s not required for long Covid. I don’t think safe to make any kind of assumption of around amillion people in the UK having ME/CFS in the absence of any evidence.