We need a diagnostic test for fatigue induced by sustained activity

The most important thing about a test might be that big pharma would be willing to get involved if they have a biomarker to measure their drugs against.

 

In which case it should be a priori clear that no test is needed to prove the realness of symptoms or disability experienced. What is needed is something that leads to an understanding of pathophysiology, in that case BPS arguments will become irrelevant.

Coming back to the topic of the thread: I think that also better PROMS are needed. I suspect that a carefully conducted long-term activity pattern monitoring study could be useful to establish those.

 

I really think FUNCAP is a step in the right direction.

 

I think so too, but of course it only captures certain aspects and other PROMS probably don't exist to capture other aspects that need to be covered. It might be interesting to see how it would do in a long-term activity tracking study, especially how up and downs can be dealt with, when it was designed to capture the average behaviour.

 

If we get a treatment, we'll need outcomes that can handle people with really significant improvements as well as minor ones. And worsening and/or new symptoms, of course.

What we're really crying out for is data gathered over time—years, not just weeks or months. Proper hospital clinics gather this, but it isn't done systematically for people with ME/CFS.

The ME/CFS and Long Covid clinics could do so much more good by diagnosing people, then saying there's no treatment but we want to follow your progress. There'd be no incentive to fudge the outcomes, and the information would be genuinely useful.

[Minor edits for sense]

 

Are there any PROMs for other conditions that have more nuance than FUNCAP for ME/CFS?

 

I think FUNCAP is about as nuanced as you can get, for measuring functional capacity in people with PEM. FUNCAP can pick up very small changes or wild changes from very severe to mild, quite well.

The problem is of course we don’t only want to measure functional capacity. For most usecases you want to measure other things like symptom burden and stuff.

 

What do you mean by more nuanced? By design it doesn't capture all aspects of ME/CFS and it might not be possible to know how much of everything it captures is meaningful when there hasn't been data to assess that. It seems to be a very well designed tool to capture some form of average impact of certain activities, but that doesn't capture ME/CFS in all its nuances.

 

I don’t think any tool can measure all aspects of something as diverse and extensive as ME/CFS.

FUNCAP measures the most debilitating aspect: PEM. Of course we can’t ignore individual symptoms, but it still measures the impact of the symptoms on your functional capacity specifically.

Level of granularity.

 

Yes it’s hilarious but not isn’t it. If it is a medical clinic then appointment might be a few times a year. Certain aspects of the blood will only change every 3months for example like red blood cells so you wouldn’t test something related to that two weeks in and expect an answer to tell you that means something has or hasn’t got worse.

I think if we think of the illness underlying of me/cfs then the change is eg at a year where you know it’s definitely a different severity and not a bout of PEM or a longer crash (because I won’t use the term relapse anymore when it’s not relapse) BUT I think ALL of those preceding months in the year leading up contributed to it eg by being over-threshold (whether PEM triggering or not, just ‘riding limits’) or not giving enough depth and length of recovery time from something more acute

and by me saying a year, I of course don’t mean it’ll go away or stay static (if continued over threshold) because that means a lower threshold comes with it.

so somehow the energy stuff needs to manage to capture nearly all that could affect us in that lead up (impact in) then see if we are as capable and unwell as we were the year before or we are a bit less due to the adaptation effect (there’s a better term a paper used for this somewhere)

if any markers do then happens to ally to either of these it feels like it would be a clue. But we don’t know where or what they’d be and the complication of course is that if the testing process involves things that cause short term over threshold then that could obscure the picture (all you are comparing is someone finding the journey to blood test easier than last time or vice versa when actually a different markers says they are more ill)

physios and OTs seem to be trained to like things like shoulder angle measurements or performance on a specific task for a stroke. And we aren’t just talking envelope size and has that changed (and do we measure that over say a fortnight or a month? And then check PEM hasn’t been triggered, how do we know it wasn’t over-threshold without a test etc?) but maybe there are other aspects. So tests that are one task where people will just compensate by showering less rigorously or being slower the next day at x fir having spent their energy on y task said physio will assess misses the point.

that’s before we get started on the issue of explaining to readers they need to be looking at each subjects’ change in envelope over that year. Vs their activity vs individual threshold. And not just thinking they’ve proved themselves because they correlated those who did less (in preceding year) with being more severe.

We have to make them care about harms and actual outcomes to do that and not slip into sophism/false science based on naff pseudo correlations being misinterpreted. And I can’t think of any other way than making a clinical responsible lifetime-wise and to have ‘5yrs outcome’ as a reporting stat by which they are judged. To make making people more severe long-term something they are as invested in as patients.

but then we have the circular what is that measured on? And how could that end up with fudging and coercion of patients to perform or say the right thing

 

But how do you assess levels of granularity when a priori it isn't designed to measure all aspects of PEM, doesn't capture cumulative efforts and hasn't been used in studies yet? It seems the relationship between PEM and cognition problems in ME/CFS seems to not have ever been studied, nor is it clear what cognitive-PEM is so it's hard to know what those impacts on functional capacity would even look like (that is not the fault of the authors that made the FUNCAP, these things are simply not known)?

On the other hand it doesn't assess for impact on sleep and it's unclear what implications that might have again and so forth. It's possible that you're measuring all of that indirectly, but who knows? Perhaps it's granual enough to detect changes happening over longer time frames but not shorter, perhaps it doesn't relate well to activity levels etc.

It definitely seems like a step in the right direction and I can only applaud everyone working on it and it certainly addresses a void and the team working on it are super cooperative and it should be used more, tested and then re-iterated but I'm unsure how one could currently know more.

 

It doesn’t measure PEM. It measures functional capacity. FC can be a proxy for some aspects of PEM, but PEM isn’t the purpose.

I feel like you’re saying that we can’t measure the granularity of a ruler because it doesn’t measure temperature.

I know the thread asked about a measure for fatigue after exertion, but I didn’t say that FUNCAP measures fatigue or PEM. So I understand why I might have caused some confusion here.

The point I was trying to make when I mentioned granularity is that PROMs are inaccurate by nature because subjective experiences are inaccurate and highly variable.

And because fatigue is subjective (at least how it’s currently defined), we will probably not be able to measure it very accurately with a PROMs. So FUNCAP is as good a measure as any.

 

Perhaps I'm the one causing the confusion, or at least I didn't mean to say that FUNCAP is intended to be a measure of fatigue or PEM.

Here is an example to illustrate what I meant with cognitive problems and functional capacity and my understanding of the FUNCAP: It seems that some pwME report that cognitive activity causes similar symptoms as physical activity whilst others report that cognitive activity causes more of an instant onset of symptoms for example they can't read anymore because their brain gets "foggy" but are physically otherwise ok. As I understand it FUNCAP wouldn't be able to measure this decrease in functional capacity as it measures the impact on other activities but not on the same activity?

To stick with your example: To know the accuracy of a ruler you have to have a metre bar or know the speed of light in vacuum or similar and I don't think we know either for ME/CFS (for example we don't know how it tracks against activity patterns). It might be hard to say anything meaningful about granularity if you can't say anything about accuracy. If you have 2 rulers one with 10 lines and another with 11 the one can be considered more granular than the other but both needed tell you anything meaningful. Would FUNCAP be considered granular enough if it would not be correlated to activity patterns at all?

I would argue the opposite: PROMs are not inaccurate by design, because subjective experiences are not inaccurate, rather than being accurate by definition. Where FUNCAP may be inaccurate is by assigning a causal relationship to a subjective, it might measure the perception accurately but the perception may not always be an accurate reflection of the situation. Highly variable might indeed by the case. What matters are that they measure something meaningful. It's possible that FUNCAP does exactly that to a certain extent for certain scenarios.

And who knows perhaps in the near future it might even be possible to objectively measure fatigue in ME/CFS (for example via eye tracking, typing speed+accuracy or some other means).

 

There are different mental tasks included, and you can answer that you have to do less the same day. Which in practice would be related to instant symptoms and the need to rest because of them.

I don’t think granularity has anything to do with what it measures. It’s just about how small differences that you can reliably detect.

Inaccurate might be the wrong word. I was thinking of how you might define fatigue as something else than what I do, so comparing differences between people is really difficult. But that’s probably covered by ‘variable’.

FUNCAP tries to solve that problem by making the score more objective. If I do X, I have to do less for Y amount of time.

I think that if we are going to design a better PROMs for fatigue due to exertion, we have to do something similar and tie it to something more objective than the subjective experience of fatigue. So we can’t measure just fatigue and hope to get more comparable results.

 

If we take them at their word. Which most people seem willing to do.

I don't, though. They simply mean different things when they say this. But this goes to what they actually believe, and it's impossible to prove either way.

But they absolutely don't believe this, and I don't care what arguments anyone can come up with about this. They are simply lying because their arguments don't make sense and they have to fake sincerity in order to make them believable.

Lying is far more common than most people think it is. It's everywhere. People lie about all sorts of things in order to get what they want. And this whole psychobehavioral "we believe the symptoms are 100% real" is total bullshit. They have a totally different interpretation of this, as different as HIV deniers who say that AIDS is real and serious, but either some oddball psychosocial construct, or a "gay plague", or whatever it is they actually believe but won't say out loud.

 

Where it does matter is at the point of (differential) diagnosis and cohorting. Re Yann04

 

Thanks all for the discussion so far, it's been really interesting to read.

In my view, it is of high importance to find a biomarker for active PEM. That would allow us to avoid inadequate surveys and questionnaires to begin with.

The problem seems to be: how do you design a study to find it? The main concerns would be lack of standardization and logistical feasibility.

You'd have to have a baseline non-PEM measurement, sequential measurements during activity, and then a measurement after a delayed period of time when the participant is in PEM.

This would probably rely on the participant's testimony whether or not they are experiencing PEM in the moment. That obviously adds in more subjectivity, but if you have a biomarker that correlates strongly with that self-report across participants, that's good evidence.

 

The main issue would be setting it up so that you end up triggering PEM in a way that is somewhat consistent across participants.

You could of course ask what normally triggers PEM for each person, but if the activity is different for each participant, you'd get a lot of pushback on whether you're actually measuring the same phenomenon between people.

The best idea I could come up with so far is 2 day CPET where you're able to follow up 3 days after the first test. This would likely limit the study to mild or mild/moderate participants, though.

You'll probably end up with some people who get PEM after this, and some who don't. Perhaps you would need to do a second tier of more intense/longer exercise for those who have more capacity.

But as long as that first study is consistent and standardized, you could do a follow up in more severe people. You would just be checking if the same marker correlates with self-reports of PEM in situations where you trigger PEM in the home.

This is just something that I'm tossing around in my head right now--would be happy to hear thoughts from others.

 

I’m thinking that you might have to start at the other end.

Measure biomarkers that are different when exercising or exerting and try to narrow it down. Not that this would be any easy task.

 

I think it would be better to find a few people with reliable PEM triggers and responses, and do PEM/non-PEM comparisons for those individuals. Comparisons between different people adds a lot more variables, complicating the data. Doing only one such person might encounter a unique response, but with more people, you can see whether the same factor changes.