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

I was going to write a long essay about why this is a misconception. But I guess I have already said most of it and I will leave it.

Maybe.

The problem is the implication that somehow publishing a lot of histories together achieves more than each one on its own. It only does if representative and this will not be. There is also a back to front approach here reminiscent of the hypermobility story. You decide that there is an illness called 'Long Covid' and ask people who think they have it to send in their symptoms. You then publish an amalgam suggesting that the responses tell us, roughly, what 'Long Covid' is. But that is backwards. What if there isn't something usefully called Long Covid or that there are three different things?

Delineating and categorising disease is a very complex and tricky business. In rheumatology we only really got to grips with our diseases after people like John Moll and Barbara Ansell re-classified everything based on careful study. Up to that point we had accounts of illness based on just collecting accounts and it turned out to prevent us from teasing out what was going on. It has nothing to do with whether it is done by doctors or patients. With are equally human and likely to mess things up. Equally both may see the need for doing things in such a way that you can rely on the findings.

Unfortunately, in general, rough accounts to not promote more assiduous and reliable accounts. They set up memes that run for decades getting more and more clogged up with misconception.

 

This seems to be where we're at with ME: the fatigue meme.

How exactly did they do this? I could find plenty of praise for them online for doing so but very little on how they did it.

And how could their methods be adapted to ME? Or would that not be possible because with ME we still have less radiological, laboratory and pathological evidence to go on?

 

I think it may have been something that was ready and waiting for observant physicians to identify in the 1970s.
Prior to the discovery of penicillin, or perhaps the improved public health after World War II, most interest in joint inflammation was in rheumatic fever - of importance both as an acute illness in children and a source of long term disability from heart failure in adults. After WWII Eric Bywaters in the UK trained physicians to take an interest in other sorts of joint inflammation like rheumatoid arthritis. Henry Kunkel was very influential in the USA, as had been Hench with his cortisone.

By the 1970s, for the first time, a few interested physicians were full time rheumatologists looking after inflammatory joint disease. I suspect that for both Moll and Ansell, the opportunity came simply from seeing large numbers of patients. Hench's cortisone had made people think inflammatory arthritis was treatable and soon they were re-trying gold and experimenting with penicillamine. Barbara Ansell was trained by Bywaters at the Hospital at Taplow. She came to see virtually all children with inflammatory arthritis in the south-east and often further afield. She was able to see that it could be divided into 6 very different types of problem. Moll worked with Wright and for some reason took a special interest in psoriatic arthropathy and spondylitis. They defined the ways these differed from other forms.

In these cases there was no population-based study involved except perhaps to the extent that for Ansell at least, once rheumatic fever had gone, all the other forms of childhood arthritis went to her - she had the entire caseload from the UK south-east because childhood arthritis is actually fairly rare.

In hindsight separating out forms of arthritis was easier than separating out types of ME. But as a simple point, Ansell (and Bywaters) documented which type of rash went with which type of arthritis (as did Moll in fact). Just listing rash is not enough.What matters is the detail.

 

So Ansell's approach was effectively a qualitative approach, with the benefit that she was able to observe all forms of inflammatory arthritis before passing judgement.

Yet is there more to it than simply attention to detail?

I note the recent NIH study on PEM, that also used a qualitative approach, yet many people here felt they did not capture their experience at all:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7530890/

If you were advising a group who was about to conduct such a study, what would you recommend they do differently?

 

This presumes that this study is the only study we will have. It's not--it's just a very early one. In the absence of the kind of longitudinal studies, it is important I think to have an idea what symptoms people are reporting, and I'm interested in the aggregate numbers. At some point going forward we will have the results of observational studies established to track this and will have data from those. Then we can further assess what these early data mean. Of course people can over-interpret a set of data and think it means something it doesn't. This is a self-selected group and the data need to be interpreted very cautiously with that and related factors in mind.

 

This underlines the value of good, open-minded clinician researchers who also happen to see a representative group of patients. I fear they're an uncommon species, at least in ME. Some of our clinician researchers have some rather fixed ideas. Others may be open-minded but probably see a selected group of patients and naturally have their ideas shaped - and likely limited - to at least some degree by that group.

It's my sense that the detail is exactly what we would want to try to get at with a large and well-designed patient survey. Or rather some leads as to where it might be worthwhile to go looking for more detail. Our discussions here on S4ME indicate that asking if somebody has fatigue is uninformative unless you also ask for details about situations when fatigue occurs/worsens. And that asking if somebody has worsening symptoms after exertion is misleading unless you also ask about what types of symptoms worsen and their time course. And that asking if somebody has flu-like symptoms is pointless unless you define what flu-like actually means.

So a first goal could simply be to refine the symptom catalogue we already have. This could include a much more detailed look at onset patterns, long-term changes to overall severity, and the frequency of remissions and relapses in long-termers.

A second goal could be to assess if there are other symptoms/symptom clusters we may have underestimated or completely missed*, most likely because they're just weird but not particularly bothersome, or because they're typically attributed to some comorbidity - but which could be the detail which provides important clues to pathology.

A third goal could be to link symptoms/symptom clusters to diagnostic criteria (you'd want to have lots of questions that allow you to judge which criteria a respondent meets rather than just asking do you meet criteria x, y or z), to severity, to time since onset, etc.

Obviously much of this would be a fishing expedition or, more politely put, the observation stage of research, and any seemingly significant finding would then need to be reformulated as a hypothesis and tested.

* I wonder how the Covid symptom study authors decided what symptoms to ask about in the first place? I don't recall there being any mention of prior qualitative studies or similar to create a symptom list? Or is it just another black hole in my memory?

 

I would recommend the hardest thing of all - to do what Alfred Brendall has done with Beethoven's music, and Pablo Casals did with Bach - to read and tell things as they really are, without attention to fashion or rhetoric.

In other words to do it well, with common sense but with inspiration.

 

All I can say, having been there for thirty years, is that however much this might seem useful, it isn't. I worked on RA without the benefit of any such longitudinal studies. Mary Corbett spent twenty years doing such a study three doors along from my office and when the final publications came out nobody noticed. The information one needs to work out what are useful ways of understanding and treating diseases do not much depend on aggregate descriptions over time. Not that progress over time is unimportant, but you know it anyway from your cases. If you don't have first hand experience of cases published papers are pretty unlikely to lead you to useful research. If we want aggregate numbers for service provision reasons or for impact on socioeconomic issues then we need population based data.

 

Yes, but I would not confuse that with the popular 'qualitative approach' as recently promoted as an alternative to proper controlled studies. Ansell herself did not venture much into speculating on causal connections but her powers of observation were vindicated by her followers, like Pat Woo, who showed that her categories had immunological and genetic bases.

 

Something that will make it difficult to recognize types of ME is the symptoms changing somewhat through the year and over time. Some of it seems to be caused by seasonal and weather changes. Some of it appears to have no apparent cause. It's all further confused by symptoms also changing in response to activity levels.

 

Two areas that seem promising for the purpose of identifying subgroups in ME appear to be the responses to CPET. Maureen Hanson showed some unpublished data a few years ago of patients having with PEM having additional unsual responses to the 2-day CPET, like for example not having the expected change increase in blood pressure on day 2.

Another areas may be changes in energy metabolism. That's another thing where we see consistent abnormalities across the ME patient population but also what looks like possibly important differences in some groups, like the finding that severe ME patients can be distinguished from nonsevere patients by the former having (I beleve) abnormal fatty acid oxidation.

The course of the illness also seems important. Some patients have an acute onset and become extremely sick but then improve and even recover completely. There is a gradual onset too, and some have chaotic symptom fluctuations while others just get progressively worse.

 

I have been thinking about finding symptoms and what to do and I realised that there is a fundamental problem underlying it all, which will come as no surprise.

I was interested in how arthritis was parsed out which, of course, then lead to treatments and better understanding of the biological processes involved. Calling everything psychological or behavioural is death to biological discovery. The idea that women are susceptible to hysteria and won't get well until they get something outside themselves to think about - women down the years have been told to have a few kids as a cure! - means that everyone with the disease loses out.

There is never going to be any doctor able to use experience of patients to put diseases into groups of observable signs and symptoms when every sign and symptom is assumed to be functional and the disease already categorised and the cure found.

Our problem is not that no one has looked at our symptoms but that doctors think they already know what is wrong with us and why we have any symptom we care to mention. They can a be explained away under the current explanation.

Longcovid is hovering on the edge of that abyss. I read this today

https://www.huffingtonpost.co.uk/entry/tips-to-beat-brain-fog_uk_5fc9144bc5b613828e32ea9f

business as usual.

It is all very depressing and I do not know what we do. I have waited 52 years for an answer. If we, and all those others in the same mess, were simply treated symptomatically and it was acknowledged that medicine does not understand what we have then there may be doctors who can do for us what was done with RA.

 

There are certainly a group of doctors who take a psychosomatic view. However, I know of many physicians who do not. They simply do not know how to start explaining ME. They accept that it is disabling but that does not help in working out how to help.

The BPS approach causes a lot of distress but it isn't the reason we don't understand ME. In the 1980s a number of physicians tried to study biomedical aspects of ME/CFS and found nothing. They genuinely tried. And as is so often pointed out there have been about 9000 papers since trying to find things. The London School of Hygiene CureME team were awarded about $2M to look again for biomarkers and came up with little or nothing.

 

Most of these 9000 papers are small poor quality studies, likely due to the BPS approach.

 

I wouldn't know but the CDC claimed they were biomedical, not BPS. I am afraid that small poor quality studies have been the rule from centres that do not take a BPS approach too!

 

Yes that's what I was saying. The illness became so marginalized for a long time that the conditions for good biomedical research simply weren't there.

At least this is a possible interpretation and believing that it's not correct means believing that BPS ideas about there being no biomedical disease, about the importance of not treating patients as if they had a disease, about the disease being easily curable with GET, and so on do not affect how society reacts to an illness and do not worsen the chances of scientific progress. That sounds completely implausible to me. Public perception of a disease seems to be very important.

Sure if, ME was an easy disease to solve, BPS ideas wouldn't have prevented progress, but when it's not an easy disease these ideas are enough to keep funding and interest at low enough levels to make progress unrealistic.

PS: and at the risk of offending, I suspect the best scientists have generally avoided ME because they could make a career in any field they wished. Why stay in a controversial funding-starved field?

 

That's probably mostly down to what Ron Davis mentioned about NIH not funding exploratory research, they expect universities to do that. Which they do in most diseases, just not in our case. The only way to get funding grants is to be super precise about what is being researched, no mapping-of-the-field allowed.

So no mapping-of-the-field ever gets done, because there are no ME (or chronic illness) departments anywhere, no possibility to apply the first step of the scientific method: observation. Digging is only allowed in places where people have dug before. Those other areas? There be dragons. Stay out.

The circumstances basically force research pathways that have little chances of yielding anything. Our researchers can only adapt to the circumstances, they can't beat them.

If we somehow got a massive amount of private funding that allowed them to work properly, I don't doubt most would take the chance to go back to basics and step back from the little nook they had to take in order for their research proposals to pass the many layers of denial and obstruction that sink most ME research proposals.

 

What annoys me most about this perspective is that it's entirely self-centered. Doctors are the front edge of the pandemic, they see it every day, they see the pain and the exhaustion of their overworked colleagues, they see the injury reports of what the virus does to some people and how it affects every single aspect of their job.

Meanwhile there is about 1/3 or so of the population behaving like nothing is going on, they don't read the news, don't know about any of this. They certainly aren't scared or stressed about any of this, they are far more annoyed at the inconvenience than anything else. They don't see any of what goes on behind the layers of security checks that have been put in place before the areas where patients are treated. They definitely aren't losing any sleep over this.

So much projection, people who cannot seem to understand that their own perspective is not universal. This is something universal to the BPS gang, they cannot seem to be able to imagine that every other person's life out there is any different from their own, they project their own self-centered perspective onto others, that if they are thinking and seeing something then surely everyone must. And everyone not only mustn't but clearly doesn't.

 

I don't think it's quite that simple. I am very close to a longhauler whose most challenging symptoms are POTs and brain fog. But it's not clear that she experiences PEM. For instance, on most days she cannot be upright for long but on some days she has walked a half mile or a mile and did not appear to experience any symptom exacerbation over the following days. So either her PEM symptoms are very subtle or have not yet developed, or she is not experiencing it. We are still watching and trying to determine if its there.