Characterizing Long COVID in an International Cohort: 7 Months of Symptoms and Their Impact, 2020, Hannah Davis et al

From the paper.
Limitations

There are several limitations to this study. First, the retrospective nature of the study exposes the possibility of recall bias. Second, as the survey was distributed in online support groups, there exists a sampling bias toward Long COVID patients who joined support groups and were active participants of the groups at the time the survey was published. Additionally, despite eight translations and inclusive outreach efforts, the demographics were strongly skewed towards English speaking (91.9%), white (85.3%), and higher socioeconomic status (see Figure S1). In future studies, more outreach and partnerships with diverse support groups, low-income communities, and communities of color can be established to counter sampling bias. Moreover, the study required respondents to have stable internet and email addresses, which may have excluded participants who lacked access and/or had low digital literacy. Lastly, the effort to complete the survey may have deterred some respondents who experienced cognitive dysfunction, or were no longer ill and did not have enough incentives to participate.

Due to these limitations, we suggest that the results laid forth be considered only in this context, and caution that extrapolation to the entire Long COVID population may not be valid.
Click to expand...
 
Jonathan Edwards said:
This study is by far the most accurate depiction of the illness I have seen, in terms of capturing the whole of the experience, despite being incomplete.

As usual I look at this through the lens of rheumatoid arthritis (RA). In my lifetime there was one study rather like this of the time course of the disease, which hoped to be representative. It was run by the very assiduous Dr Mary Corbett and continued for twenty years. Sadly, it made no contribution to the progress that we made with understanding the disease. Why? Because every physician involved in the disease already knew the histories of about 500 individual cases. Lumping them together seemed to have the stamp of being quantitative but that was not actually what was needed and turned out to be skewed hopelessly by predictable things like loss to follow up.

For RA the plot for 'vision' would be a tiny streak at the bottom. But this would represent the very few patients with devastating complications like scleromalacia perforans.

I could continue, but this is vey unlikely to be an accurate way to delineate individual patient experience - the way to do that is for a physician to sit down and listen a hundred times. I got fascinated by RA as a student and did that and I never needed to do more in order to know what I needed to do in the lab to explain things. Numbers do not necessarily give an accurate picture. Where you need numbers, like genetic and disease burden studies it is imperative to collect them in a reliable way.
Click to expand...

The problem is that in ME there are no physicians involved who have seen about 500 individual cases. There are some, but their experience is dismissed by the "experts" who talk about another disease entirely even if they call it ME. I have never recognised myself or my experience in any of the descriptions I come across in 90% or more of published papers. The whole emphasis is on fatigue whereas you could take away all the fatigue I experience and it would make a trivial difference to my life.

I go to see a doctor or consultant and even if they have any knowledge of ME they have no insight into my life the way they would if I had MS or RA. They assume I am in a wheelchair because I am too tired to walk even if they do not believe I am mentally ill whereas I can't walk more than a few steps because I have little control over where my legs go and because I have dysautonomia (not comorbid, part of the syndrome) and get faint if I stand. So trying to walk causes falls and pain in my legs not to mention a post exertional syndrome a few days later with sore throats and swollen glands and inability to do the activities of daily living.

If this sort of research is not the proper way to start where do we go? You are a genuine resource to our community in giving us an insight into how doctors think. What can we do that would be acceptable to them and that would move us on from the faulty narrative that ME, fatigue and tired all the time are just different points on a single spectrum.

I find it breathtaking that the report of the Royal Free epidemic and now reports of longcovid match my symptoms much, much more closely than the description used to decide if I need benefits or not. The disconnect sends us straight into MUS territory.

Feeling fatigue is common to so many things that doctors think they have seen 500 patients with ME but they have not (or they would know how common it is for us to get neurological problems and swollen glands and all the other things and there would be a mention somewhere)

If asking patients to list symptoms is not the way to go what is? (Not rhetorical, it is a genuine need in our community)
 
Mithriel said:
If asking patients to list symptoms is not the way to go what is?
Click to expand...

My point is simply that asking patients to list symptoms is exactly the right thing to do and is the basis of all clinical assessment. But averaging out a lot of people's lists on graphs does not make the information any more useful. If anything is confuses things. That is why I always like to see raw scatter plots and individual time curves for every case in published data.
 
Jonathan Edwards said:
This study is by far the most accurate depiction of the illness I have seen, in terms of capturing the whole of the experience, despite being incomplete.

But, with respect, we do not know that and it probably isn't.
Each panel seems to show an average time profile and that is not the reality. The reality is each individual case - some might go up and some down. Note that relapses and remissions seems have disappeared completely from the plots.

As usual I look at this through the lens of rheumatoid arthritis (RA). In my lifetime there was one study rather like this of the time course of the disease, which hoped to be representative. It was run by the very assiduous Dr Mary Corbett and continued for twenty years. Sadly, it made no contribution to the progress that we made with understanding the disease. Why? Because every physician involved in the disease already knew the histories of about 500 individual cases. Lumping them together seemed to have the stamp of being quantitative but that was not actually what was needed and turned out to be skewed hopelessly by predictable things like loss to follow up.

For RA the plot for 'vision' would be a tiny streak at the bottom. But this would represent the very few patients with devastating complications like scleromalacia perforans.

I could continue, but this is vey unlikely to be an accurate way to delineate individual patient experience - the way to do that is for a physician to sit down and listen a hundred times. I got fascinated by RA as a student and did that and I never needed to do more in order to know what I needed to do in the lab to explain things. Numbers do not necessarily give an accurate picture. Where you need numbers, like genetic and disease burden studies it is imperative to collect them in a reliable way.
Click to expand...
I understand your perspective as a physician, from the disease side of things. But illness is a fully subjective experience, this one especially so because it cannot be summed up in a single sentence or feature. That sets it apart from most, although definitely RA is closer to this set of problems than most. But many rheumatologists are pretty hostile to us so that's not sufficient to actually grok things, to have an intuitive understanding beyond what can be gleaned at a distance.

But the sitting down and listening to hundreds (well, thousands by now) of patients is what I have done over the last few months, every day reading dozens, adding on to years of doing the same with patient accounts of ME. Not with an understanding of disease, I have none and understand almost none of it. But I do understand the experience of the illness. It doesn't all fit my subjective experience but after reading so many stories, the patterns are obvious, the same way I can recognize a musical style even through vastly different songs. Or like those people who work at poultry factories and can tell male from female chicks even though they can't explain how. But they can, they rarely miss.

This really is the closest account of the overall illness experience so far. Obviously: so far, there will be better in the future, or I dearly hope so. And it completely misses the disease aspect but this is what we have to work with. But as an account of illness, this is so far beyond everything that has been done before it's almost equivalent to the invention of the microscope and finally being able to see microbes with our own eyes. Or adjusting the antenna to turn the static mess of a distorted picture into crisp UHD signal. There have been a few good papers so far but they have all missed most of the illness. But the results do mostly align, even when restricted to positive tests or other more stringent criteria.

I'm well aware this is of limited usefulness to physicians and the traditional disease-first approach, but I frankly don't really understand the thought process of how physicians are supposed to recognize illness as recounted by patients if they are not aware of how things unfold before the patient gets to the clinic. Because this is how the patients will explain it. If the story isn't familiar it's impossible to compare to it, as we see with all physicians coming down with Long Covid who cannot possibly imagine this has anything to do with "chronic fatigue", which yeah, we know. This is not something that can be guessed, the only way is to interrogate and analyze rigorously. They can be treated as hostile witnesses but the investigators have to record every piece of evidence and treat it as potentially useful. Insisting on disease and only disease is how things failed us. It usually works very well but not here, not for illness, not for understanding symptoms.
 
Jonathan Edwards said:
My point is simply that asking patients to list symptoms is exactly the right thing to do and is the basis of all clinical assessment. But averaging out a lot of people's lists on graphs does not make the information any more useful. If anything is confuses things. That is why I always like to see raw scatter plots and individual time curves for every case in published data.
Click to expand...
Oh no question about this, fully agree. Aggregate data only gives us the widest trends, a map of the field where digging should be planned for. This needs so much granularity, depth. But no one has ever done this first step before, making sure we have data integrity before we start reducing to subsets. This is a first step, an excellent one at that, one that was never made for us, failure of which lead us down a path where a complex wave function with a high information density is reduced to a single binary digit, losing 99%+ of the information the wave contained. This is not the way to work. This may be the worst possible way to work.

I cringe about this as an IT specialist. Erasing information, either by accident or purpose, has about the same place in my profession as do no harm does to medicine. Sometimes it happens, but when it does people do get blamed and mistakes aren't repeated because losing information when the entire point is to gain knowledge from it is problematic. And when it does happen there are usually contingency plans available anyway. Here decades of relevant information were simply wiped out because the recorder wasn't even turned on. In my profession that would get everyone fired, definitely no one would get a knighthood or be awarded specifically for being so effective at erasing data.
 
rvallee said:
But no one has ever done this first step before, making sure we have data integrity before we start reducing to subsets.
Click to expand...

But this is not 'data integrity'. It would be great if it was, but it is almost certainly grossly biased and obfuscated, simply because data gathering almost always brings those problems with it unless you work hard to avoid them.

I think I understand the illness of ME because I have sat and read what a couple of thousand people on forums have described in great detail. If I tried to put that in to some graphs I would produce a lemon. For people wanting to work on Covid-19 or ME I suspect those graphs are actually pretty meaningless. There seem to be some upswings and some downswings that might mean something but equally they may be artefacts of acquisition. I am getting a reasonable feel for the problems of Covid-19 from personal accounts through media. Without those the graphs would mean nothing to me and I am not convinced they tele anything more to be honest.

For sure, it raises the profile of the problem, but the authors response to the limitations of their study seems to be to do more studies seeking out ethnic minorities etc. From my experience there is nothing to be gained from more studies like this and if you want to avoid bias you need to do prospective population studies.
 
Jonathan Edwards said:
My point is simply that asking patients to list symptoms is exactly the right thing to do and is the basis of all clinical assessment. But averaging out a lot of people's lists on graphs does not make the information any more useful. If anything is confuses things. That is why I always like to see raw scatter plots and individual time curves for every case in published data.
Click to expand...

Thank you, that is what I mean by saying that your experience is a resource for us.

One of the reasons we are stuck with fatigue is that the "experts" said that most people agreed they experienced fatigue (can't remember where) but they did not ask anyone if it was the most serious or disabling symptom for them or if they considered it to be very important. But this gave the BPS crowd the justification for introducing CFS and that is what was wanted.

A moment's thought shows how stupid that was. Everyone with TB fees fatigue but putting it on a spectrum with other fatiguing illnesses tells you nothing about how to treat it.

We need to debate how to get across exactly how ME affects people.
 
Jim001 said:
Is anyone else a little surprised to see the mast cell related charts not showing more prevalence? I'm looking at the anaphylaxis and allergy charts. They're pretty flat. I dont know what the prevalence of mast cell disorder is in the ME population but I would think the majority experience it to some degree, whereas in LC it appears to be very low.
Click to expand...

There is no evidence that mast cell disorder is very common in the ME population. I am not surprised by the charts at all.
 
Last edited by a moderator:
Snow Leopard said:
This study is only intended to be of "suggestive" quality. The primary purpose is to attract attention and justification for funding higher quality studies.

If the medical research community did their job and initiated prospective population based studies from the start of the pandemic to study long term post-viral outcomes*, this patient-driven survey would never have been needed.
Click to expand...

I agree. It doesn't "prove" anything but provides a lot of rich data that will hopefully provide helpful leads for future research. No one should be claiming this research is anything other than what it is--a non-randomized survey providing qualitative and anecdotal data from those who respond, for whatever complex set of reasons. It alerts people there appears to be a real issue here, although it cannot be relied on to provide reliable information on prevalence, average length, and other population-based metrics. It shouldn't be criticized for not being what it's not intended to be, unless the authors or others are promoting it that way or making overstated claims.
 
Last edited:
dave30th said:
That's interesting. Do others experience a similar split between mental PEM and physical PEM if they've engaged in only mental exertion?
Click to expand...
Well, obviously without asking that question in this kind of study it would be difficult to clearly get a grasp on how many do or don't but I personally don't, and I don't see many others talking about this kind of split either. (Which is in no way trying to claim that it doesn't happen, just that it doesn't seem to happen for the majority of patients who talk about such things.)

dave30th said:
I agree. It doesn't "prove" anything but provides a lot of rich data that will hopefully provide helpful leads for future research. No one should be claiming this research is anything other than what it is--a non-randomized survey providing qualitative and anecdotal data from those who respond, for whatever complex set of reasons. It alerts people there appears to be a real issue here, although it cannot be relied on to provide reliable information on prevalence, average length, and other population-based metrics. It shouldn't be criticized for being what it's not intended to be, unless the authors or others are promoting it that way or making overstated claims.
Click to expand...
Exactly, which is why I don't understand the criticisms of the study, or at least the criticisms of the study seem to be based on an assumption that the study is being misrepresented in some way and that doesn't seem to be the case.

I very much hope that, as I've said above, that we can do something similar for ME. At the very least, it would provide a way that a far larger portion of the patient community could make their own voices heard than would be possible if we had to rely on the medical community listening to our individual stories.
 
Andy said:
Exactly, which is why I don't understand the criticisms of the study, or at least the criticisms of the study seem to be based on an assumption that the study is being misrepresented in some way and that doesn't seem to be the case.

I very much hope that, as I've said above, that we can do something similar for ME. At the very least, it would provide a way that a far larger portion of the patient community could make their own voices heard than would be possible if we had to rely on the medical community listening to our individual stories.
Click to expand...
My impression is that it's an illness-first approach, treating symptoms with respect rather than a neutral feature of no meaning used only for diagnostics and which can be ignored other than as an annoyance afterward.

Medicine considers the patient experience to be tainted, unreliable, unworthy of paying attention to, to the point of being openly hostile to it. Therefore an illness-first approach is invalid because medicine is strictly about disease, illness and symptoms are for patients and nurses to worry about. Symptoms don't kill people and only death matters, or at least it matters 100x more than every other outcome combined. Some will be offended by the suggestion but in practice it's sadly true.

There is nothing wrong with this study besides the fact that it is not how medicine normally functions. This normal function of medicine has been a failure with us, however medicine cannot acknowledge that for reasons that escape rational thought and basic ethics. So it must be wrong because it's not how things usually work. Even though those normal things don't work, but no one wants to acknowledge that. Infinite tolerance for failure is better than admitting failure, somehow. Even though it harms, but what really is harm, if it comes from the patient perspective, it's wrong. QED.

It's the equivalent of "how dare you suggest my hands are dirty before I plow them raw into my patients' wounds". It's wrong because it's different and suggests the physician is at fault for doing it wrong. Which is true.

Which frankly more than anything tells me that my suspicion that medicine wants nothing to do with patient engagement to be even more pathological than I thought. Medicine wants none of this, absolutely none. This is sadly more evidence of it. So much to change before we can make progress.
 
dave30th said:
That's interesting. Do others experience a similar split between mental PEM and physical PEM if they've engaged in only mental exertion?
Click to expand...

Interestingly enough, some people do and some people don't. There have been several threads discussing how PEM is experienced, which I'm having trouble finding because "PEM" is to short to be used as a search term in titles.

I've found the most recent one though:

https://www.s4me.info/threads/poll-physical-vs-cognitive-pem-same-or-different.16948/

The poll question was:

"Please compare your typical PEM from physical overexertion with your typical PEM from cognitive overexertion. Consider the typical range of symptoms you get in each case, plus their timing and severity. Do you feel both types of PEM are fundamentally the same process or do they feel fundamentally different?"

The most common answer was:
"Some aspects feel fundamentally the same, but others feel different."

But 21% reported:
"They feel fundamentally the same"

The sample size was small, so make of that what you will...
 
Mental exertion makes the cognitive parts of my brain shut down. I can't think or process what I see, my speech goes and it is difficult to move immediately.

It is different from physically doing too much but in my experience it is the same processes. If I use my legs too much, I get pain there and it is near impossible to walk for instance. Some of the difference is that pain stops me using my legs to the point of shutting down but with mental exertion it is easier to go to an extreme.

But doing too much of either can trigger the 3 day delayed PEM which is similar in form for both.
 
You must log in or register to reply here.
Back
Top Bottom