Developing a blood cell-based diagnostic test for ME/CFS using peripheral blood mononuclear cells, 2023, Xu, Morten et al

Yes, but you answer this one with your number 2. There is no such thing as 'accuracy of an ME diagnosis'. It is merely a placeholder for categorising your illness which very likely falls within a complex mix of overlapping problems - as for RA. I never bothered too much about the accuracy of a diagnosis of RA. I tried to work out what was wrong for that individual within a spectrum problems.

But don't you see the contradiction here, Robert? If the correlation is 100% - high specificity and sensitivity - then the predictability is exactly the same as the clinical picture - by definition. As a maths teacher Trish should agree with me here!

Of course the test might be more reliable in GPs hands than an inexperienced clinical assessment but if we are talking about research we are comparing it with skilled assessment.

Your number four is probably right very often but if the test just reflects eating burden or lying in bed then you come unstuck.

I agree that a good test will reduce stigma but what I am saying is that good tests in medicine do that because they tells something other than clinical assessment. In real practice tests with sensitivities around 80% are often ideal. Specificity in the general population is often not very important.What is much more relevant is specificity amongst a group of people who for one reason or other might be considered for the diagnosis. So the specificity of rheumatoid factor in people with soft joint swelling is much more relevant than specificity in the general population.

But sensitivities of 80% mean that you have to make sure you do the right thing for the people who really do have X wrong but are test negative. And that is why rheumatoid factor is a very bad test in general practice because it is negative quite a significant proportion of RA cases, particularly early on.

 

Well, sure, and isn't that the point? To assist clinical judgment?

As for predictive value, I'm sorry, but reliably predicting what is going to happen isn't necessarily part of the deal., e.g. many pathogen-based diseases, particularly in acute cases, but more contentiously in late or post acute cases. Depending on the pathogen, there are yes/no binary diagnostics. For instance I can direct culture for spirochetes, and if I get a hit, that positive tells me I'm infected, but it carries almost zero reliable predictive value as to the infection's course. There are simply too many variables at play.

That would be nice, of course, but not required.

 

If a test couldn’t differentiate between someone with ME/CFS and someone who had similar activity levels for a different reason, I wouldn’t consider it to be a diagnostic test – just as actigraphy is not a diagnostic test.

 

Yes, but sensitivities and specificities on machine learning against healthy controls are not going to tell you that. You could compare against inactive people but if you do not know what the confounding association is you won't know how to tell if the test is meaningful.

It is all very complicate. Take a test like urate level for gout. Plenty of people without gout have urate levels above the 95th centile. People with gout often have rates below the 95th centile. But if you understand what matters is the crystallisation threshold, not the centile, and that people with gout are rarely much below the threshold, taken together with other evidence you can make good use of the test.

 

I am "sure" of the clinical diagnosis when a team of ME/CFS experts screening for a study is involved, but those are very hard to come by so for everyone else a biomarker would be quite useful. In ordinary clinical practice doctors don't have the faintest idea of what ME/CFS even is, much less how to diagnose it correctly.

 

There is the issue of how confident one can really be about 'expert' diagnosis. In fibromyalgia one study showed a hundred fold discrepancy in the diagnosis rate in 'expert' rheumatologists. But let's accept that ME isn' that hard to diagnose it you think is real.

I appreciate that in a perfect scenario what you say seems to make sense.But I wonder if you should be careful what you wish for.Think of the following scenarios:

Lab scientist bounds into Dr Wak's office:"Hey Wak, you know what, in 100% of your ME samples and none of the controls the samples go green if we add chemical XP789D5." "Gee, Moncado, that's amazing. Have you no idea why?" "No clue, Wak, I can't make any sense of it at all, it does not seem to indicate anything that would explain the symptoms." "Ah, that is really interesting. So maybe this is a cool way to pick out those lazy people who have false beliefs about having some strange disease, when in fact there is nothing to suggest they do. Right, eh?. The Disability folks could save billions on this. We could identify all the lazy guys with a simple blood test."

Alternative scenario: :"Hi Dr Wak, I found something I think might be interesting. "What's that Susie?" "Well it is only half the patients but they have this antibody that suggests that they have a problem with gut immune cells." "Amazing, Susie, so at least we know that in some cases this is likely a real disease. And I guess the others probably are too. We will need to train up a whole new set of GI physicians to treat ME seriously and find out what is wrong in the other half."

 

Agreed. I suspect that irritatingly, if we then did have the "ideal biomarker that informs the mechanism of the disease", the clinical acumen would substantially increase overnight. Doctors would start taking a history and listening to the patient in the context of this understood mechanism.

Eg if the biomarker were to clearly show that the mechanism involved a specific acquired mitochondrial defect and that energy generation on demand was severely impaired, with an in-built delay from this inadequately met energy demand causing downstream physiological derangement and symptoms: then suddenly doctors would be enquiring about the unique symptom of PEM and diagnosing patients accurately.

The confirmatory biomarker test might then be optional (esp if it was high cost, low availability). The pre-test probability from the unique symptom of PEM could well be high enough on history alone. Which is of course something we could have right now if doctors —

 

wow you just turned my entire thinking on its head. I'd always thought 'if only something would show abnormal on a test then people would believe me'. But sh*t, its not that simple is it.

God, can you imagine, the thought of that scenario makes me tear up. Its like a wild fantasy.

But i can see that it would have to be something that explains the mechanism for symptoms or at least points to it. Makes me a bit scared of Decode TBH. We dont want
"Lazy gene identified" to be the headline!

 

Absolutely. To us it seems easy to diagnose this but the average doctor out there is clueless and quite often hostile.

 

Even this happens, it doesn't matter. Once the patients are actually diagnosed, we can leverage on huge numbers and make stuff happen with advocacy, instead of having 90% of the patients undiagnosed and the other 10% chasing alt med bullshit because even to them the illness doesn't feel like it's real.

If we have unity, we can overcome any obstacle. The problem is right now there is no unity because the patients are not being diagnosed or they are convinced by doctors that cfs is "something else". And we have no credibility to convince them otherwise because there is no objective marker.

With an objective test, we can credibly claim that this really is a scientific mystery that needs to be researched more, instead of some mentally ill patients/grifter researchers roleplaying their involvement in a real illness. Many doctors would still be against us but it doesn't matter because we could overpower them. We could also try to get top scientists involved from physics, bioengineering backgrounds and so on.

The problem with our advocacy is not that we have enemies, everyone does. The problem is that we don't have a leg to stand on and the whole thing from the outside (or sometimes even from the inside) looks like it's not real. Having a biomarker would help us immensely in this regard.

 

Wouldn't we also expect that if you take any general cohort of people who have been diagnosed with ME or CFS, you're going to get a lot of people who probably do have something else? Maybe ME is more than one thing or maybe you have a bunch of people misdiagnosed, but I'd be immediately suspicious of any test that doesn't give you a significant number of false negatives.

Enabling researchers to study a homogenous cohort of people with ME by identifying them with some kind of test would seem like a pretty useful step, even if it isn't used in the clinic. That said, I don't think this is that test.

 

Yes, this is one of the points I was hoping to get clear.
Except that it probably won't be a test for 'it' - i.e. ME.
ME, like RA is a useful clinical category. The tests that are useful in RA are tests for various aspects of processes that we can intelligently target. 'RA' comes to be an umbrella for a Venn diagram of 'its' that the tests are accurate for.

 

The (ongoing) Aberdeen University study on fatigue , i.e. using actimetry, could potentially help Doctors? Currently doesn't include ME though!

 

Hey did you consider contacting the study authors regarding this?

Thanks

 

Heh, not likely. Disability is so hard to get and pays so very little that actual fraud motivated by 'laziness', or anything that could be described as such, is so small that they'd actually lose money if everyone was tested for it, because it's just that small. And we're far from the only controversial medical condition out there.

Disability fraud is one of those things that has been massively hyped but is actually very rare. It's similar to how so much is written about retail theft, when actually wage theft is 10-30x as large, and no one ever talks about it. Because the press caters to its owners, not its readers.

Disability systems are built to be punitive on purpose: so that they aren't even worth exploiting. It's so much easier to just do light crime, work under the table, do petty scams or theft, etc. And still they are quite massive but that's because there is a lot of disability out there, and medicine's inability to process it has sort of caused a stalemate where the problem never gets any better.

In large part, the growing acceptance of 'mental illness', used loosely, is mostly not because it is more accepted, but because of that strategy of simply throwing any unsolved problem into this other giant pile of unsolved problems. Mental illness is no more or better accepted that it was 30 years.

It is talked about more, sure, but effective help hasn't actually manifested, in part because of this weird obsession people have about being cheated in ways that don't even make sense, and because of an entire industry built around ineffective solutions to what are mostly mislabeled problems. In a sense, this is similar to the return-to-the-office movement happening even though this way is actually less efficient and productive. It's not actually about productivity, it's about control.

Disability systems know this, they actually count it. They genuinely don't care about testing people for laziness, they know it's not a real problem. In fact they know that most disabled people can't even get disability, the system is built this way on purpose.

Of course they will save a lot of money by actually treating us, but hardly anyone seems capable of that kind of reasoning. Oddly enough. Obsessing over made-up problems while ignoring real solvable ones, now that is the human way.

 

Of course, I was just fantasising what Dr Wak might say if he believed everyone with ME was lazy - and of course there are plenty of Dr Wak's of that sort around.

 

Heh, wasn't sure but thought so. I made the comment anyway because it's definitely a thing that comes up a lot, and why even though there is so much frustration going around about 'lazy diagnoses' making it easy to get free money, disability systems are really not likely to be bothered to do something here because they know better, they're just built this way. I thought that Long Covid would sort of kick that into motion but it really turns out that those systems aren't built to solve their own issues.

It's really hard to see who will even put institutional weight behind this. I saw today that the UK government is making this big push to get therapists to convince the long-term ill to get back to work, since the NHS can't help them. So they're spending money that won't do anything because they don't want to spend money that would do something, if it was well-spent. It's all so bizarre.

Really not a lot of sense going around.

 

Good thought, I might come back to this if and when my health is on more of a relative high