Epidemiology of ME and CFS

In conjunction with the master list of facts @JaimeS is discussing in this thread (https://www.s4me.info/threads/a-masterlist-of-me-facts-the-citations-that-support-them.5315/unread), I would love help improving MEpedia's epidemiology page: https://www.me-pedia.org/wiki/Epidemiology_of_myalgic_encephalomyelitis_and_chronic_fatigue_syndrome

I was working my way through every published epidemiological article but lost steam!

I think a good first step would be to complete the "Incidence and prevalence by country" and "
Incidence and prevalence by definition" tables to make sure they are complete and include every study (including those using definitions that might be considered poor).

Might someone help with this? Or do you have missing studies to suggest?

 

I don't know if that's the best way to go forward, since there are very few qualitative studies on epidemiology. I would focus on the major ones (Jason et al. 1999, Reyes et al. 2003, and perhaps also Nacul 2011) and reference the other ones as extra information (or even a curiosum), in a different heading.

I'm from Belgium, a country where ME/CFS research and advocacy is very much alive, but there is not a single epidemiological study available here. The situation is not much better in The Netherlands. Here they reference to two 20-year old surveys (https://www.ntvg.nl/system/files/publications/1997115230001a.pdf and https://www.ntvg.nl/system/files/publications/1997115200001a.pdf) where GP's had to report their cases of CFS (one study used the Holmes criteria, the other one created its own definition). Obviously this doesn't work, since GP's don't know much about this condition (certainly back in 1997). The result is a serious underestimation of the prevalence rate (both surveys reported approximately 0,11%, instead of the 0,42% Jason found in the US).

So I’m afraid this kind of ‘information’ causes more confusion than insight. Off course those studies should be cited and linked to, but I wouldn’t present them as reliable information.

PS: a large UK biobank analysis also reported a prevalence around 0,4%: https://mecfsresearchreview.me/2018...emonstrates-an-inherited-component-to-me-cfs/

 

Fact sheets are for pulling only a single figure or range based on the highest quality data. This page offers us a space to be complete and transparent re: where these numbers come from.

And re: the "by country" part, we already do that. We say 250,000 UK and 1 million US all the time. So if those figures are being commonly used, it's worth citing where those estimates come from.

I also think it's important to show how widely the prevalence rates change depending on which criteria you use. There may also be specific instances where people from the UK will need / only want to cite UK studies, US, US studies, etc.

 

Perhaps in the introduction you could make the point that the estimated prevalence (using the Fukuda-criteria) is somewhere between 0,2%-0,4%, cite the major studies, and then later show the wide range of % reported by others and the different rates per country. I'm concerned that if you present them all in one table, this might be misguiding since many of the other studies are not very reliable (at least in my opinion).

Take the case of Belgium or The Netherlands. In the famous 2018 report of the Dutch Health Council it is stated that there are about 30.000-40.000 patients in the Netherlands. Someone working on the MEpedia page might take this figure, put it in a table for the country 'The Netherlands' and cite the report. But those figures are unreliable because they are partly based on the two surveys I cited in my previous post. The same is true for Belgium, where government reports use the Dutch prevalence and adjust it to the Belgium population size. (What they actually did was: they noticed that the 0,11% figure was off mark, and then they opted for the low end of the US-based studies: around 0,2%)

Just hope the text will somehow make clear that those large epidemiological studies (Jason et al. 1999, Reyes et al. 2003, and perhaps also Nacul 2011) are more reliable than others.

Part of advocacy here is to challenge the figures the government proclames and plead for the more reliable estimates reported by large US/UK population-based studies (which are almost 2x as high).

I think those figures are cited a lot because they are easy to remember. 250.000 for a UK population of 65 million, is a prevalence of around 0,4% (It's actually more complicated because you have to adjust for the non-adult population). And I think the 1 million figure is just the 0,42% prevalence rate of Jason et al.1999, that was cited years ago, when the US population was a bit smaller (but the figure stuck). Please correct me if i'm wrong about this.

 

Yes I think that's exactly what we should do. Those should be what comprise the first few sentences of the incidence/prevalence section, before the table.

There's already a sentence there. It just needs to be expanded. Want to jump in?

Estimated incidence rates – generally, the number of new cases in a single year – vary from 0.025%[1] to 0.3% of the population. Prevalence rates range from X to Y. This variation is due in part to the variety of definitions used and sampling methods.​

 

Prevalence rates range from X to Y. This variation is due in part to the variety of definitionsused and sampling methods.
This suggests you’re going to use the lowest prevalence in all studies as X and the highest in all studies as Y. The CDC once estimated the prevalence to be 0,01% or even lower (https://www.ncbi.nlm.nih.gov/pubmed/8387910) I don’t think we should use those, except in a history section.

I would write something like: “the prevalence is projected at 0.2%-0.4% (cite important studies), though estimates vary widely due to different definitions and sampling methods used (cite review for example: https://www.ncbi.nlm.nih.gov/pubmed/23576883).

Sorry for making all those critical remarks without actually helping much. Really appreciate all the hard work you do.

I prefer writing a ME-pedia-page after I've read all the material throughly (and I haven't in epidemiology, will take me a long time to do so). But I will try to add the information about the Dutch surveys...

 

Thanks! Why don't you go and make the change you suggested? Happy to help with any technical questions. I understand wanting to read thoroughly (and it's also much more satisfying!) but what you are proposing is already an improvement over what's there, so I think you should implement the change. It can always be further improved later. This is an iterative process.

 

Just an observation and I think it was Osler’s Web Hillary Johnson who picked up on it, the figure for ME sufferers in UK is oft quoted as 250, 000 and has been so for decades, it never changes...I wish I could recall figure in 1980s but am afraid I can’t. I just wonder how reliable these figures are, especially given criteria issues....

 

Also, was just having a peek at the MEpedia pages. I don’t know - is it actually the case to say that MCAD and ME are commonly comorbid? Do we know that for certain? Also, not sure if I agree that ‘The symptoms of MCAD can also be very similar to that of myalgic encephalomyelitis (ME) and therefore may be confused.’

I myself have issues with sensitivities that cause chronic sinusitis - these symptoms developed a long, long time after ME diagnosis - but this is def not same as my major ME brain/muscle symptoms, though I accept the sinusitis may be part and parcel of immune system in disarray.

 

We don't and it shouldn't be there. Afaik, there are no studies looking at mast cell activation in ME(except one, maybe), although there should be. Having mast cell symptoms doesn't mean that it's MCAD. Mast cells are constantly reacting to the environment so it wouldn't be surprising that they are secreting in ME, but that doesn't mean that it's MCAD and such speculations, I think, belong to forums and groups, not MEpedia. At least for now.

Here's an excerpt from a recent (2017) paper by Cem Akin on MCAS (https://www.jacionline.org/article/S0091-6749(17)31025-4/fulltext):

"AREAS OF UNCERTAINTY AND OPPORTUNITIES
FOR RESEARCH
In clinical practice some patients with a variety of multisystem
symptoms who do not have an identifiable central cause for their
complaints are referred for investigation of mast cell activation
syndrome. These symptoms can include chronic fatigue;
intolerances to various environmental factors, foods, and
medications; and neuropsychiatric findings, including memory
problems and headaches. These complaints can be present on a
chronic basis without well-defined attacks or episodes of mast cell
activation. Currently, there is no evidence to suggest that an
abnormal mast cell phenotype that results in ongoing chronic
mediator release is responsible for these symptoms. Some of these
patients can have a slightly increased basal tryptase level that
might have led to the diagnosis of mast cell activation. Familial
hypertryptasemia should be strongly considered in these patients
because its prevalence in the general population appears to be as
high as 6%. There are also clinical observations of patients who
present with hypermobility-type Ehlers-Danlos syndrome and
postural orthostatic hypotension who also have various symptoms
of mast cell activation, such as flushing and gastrointestinal
complaints.48,49 A subset of patients with hyperadrenergic
postural orthostatic tachycardia syndrome were reported to
present with increased urinary histamine metabolites and are
more likely to experience flushing, shortness of breath,
headaches, diuresis, and gastrointestinal symptoms.50 More
research is clearly needed in these areas because it is not clear
whether symptoms attributable to mast cell activation in these
patients result from mast cell mediator release or are caused by
another pathologic process, such as dysautonomia, defective
connective tissue, or both (Table VI)."

 

As @Nasim Marie Jafry correctly points out, incidence and prevalence are not static. It's therefore important to state when (and where) the data were collected and analysed. I suspect also that there is a lot of extrapolation from small datasets.

Just because a number has been bandied about for many years doesn't make it accurate (or true).

 

My feeling too, Pechius, is the MCADS/ME connection is in its infancy of being researched and we should not be documenting a comorbidity as confidently as MEpedia has done, when it is perhaps not true.