-
New Forum Software Installed (Still a few issues but reopening the forum) More details.
Fatigue outcomes following coronavirus or influenza virus infection: A systematic review and meta-analysis : 2020 Chalder et al
- Thread starter Sly Saint
- Start date
Woolie
Senior Member
I'm honestly really not sure sometimes if this team are having genuine trouble getting their heads around the difference between correlation and causation, or whether they fully understand these concepts, but pretend not to so that they can draw conclusions that are in line with their beliefs - which seem to be to do with the perpetuating role of "psychological" factors in post-viral fatigue and/or the value of their own interventions in addressing them.
Obviously, those psychiatric diagnoses are not sound evidence of any casual relation. In some instances, they are confounded with the illness itself (for example, fatigue alone will increase your scores on some depression self-report scales), in others, they are most readily understood as the distressing consequences on losing one's health and life.
The authors make a big deal of the fact that some of the studies are - in their terms - "prospective", which simply means they measured the same variables acutely and post-acutely. But since we already know there's a link between the severity of the acute illness and the likelihood of developing long-term fatigue, this illness severity factor confounds the interpretation of pretty much all other variables, because all are likely to be associated with illness severity.
The kind of prospective cohort study you would need to demonstrate causal relations between psychological factors and fatigue would need to measure people before they first got sick, and show that certain psychological factors were associated with an increase likelihood of ongoing fatigue following the acute illness.
Some of the factors these authors mention are actually laughably non-casual - how likely is it that the fact of being admitted to an ICU acutely actually makes you more prone to long-term fatigue? That's as hilarious as the idea that being admitted to hospital makes you more prone to death! Obviously, this measure is picking out something about severity in the acute phase that predicts the degree of ongoing fatigue. But can they not see that the measures of anxiety, etc, may be doing exactly the same?
Ditto for taking more time off work and having decreased coping. It just seems laughable to say these things give you fatigue, when the far simpler interpretation is that these are a consequence of the ongoing fatigue.
This is a not a sound conclusion from the data, because it makes a causal claim which the data do not support. What the various studies report is that there is an increased incidence of anxiety, depression and psychiatric diagnoses in the samples with significant ongoing fatigue relative to controls. Sometimes, this is even present in the acute phase. Ongoing fatigue is also associated with increased length of acute stay in hospital, ICU admission, and also with reduced work hours, reduced exercise capacity, reduced personal coping, increased cognitive problems, increased medicolegal involvement.
Obviously, those psychiatric diagnoses are not sound evidence of any casual relation. In some instances, they are confounded with the illness itself (for example, fatigue alone will increase your scores on some depression self-report scales), in others, they are most readily understood as the distressing consequences on losing one's health and life.
The authors make a big deal of the fact that some of the studies are - in their terms - "prospective", which simply means they measured the same variables acutely and post-acutely. But since we already know there's a link between the severity of the acute illness and the likelihood of developing long-term fatigue, this illness severity factor confounds the interpretation of pretty much all other variables, because all are likely to be associated with illness severity.
The kind of prospective cohort study you would need to demonstrate causal relations between psychological factors and fatigue would need to measure people before they first got sick, and show that certain psychological factors were associated with an increase likelihood of ongoing fatigue following the acute illness.
Some of the factors these authors mention are actually laughably non-casual - how likely is it that the fact of being admitted to an ICU acutely actually makes you more prone to long-term fatigue? That's as hilarious as the idea that being admitted to hospital makes you more prone to death! Obviously, this measure is picking out something about severity in the acute phase that predicts the degree of ongoing fatigue. But can they not see that the measures of anxiety, etc, may be doing exactly the same?
Ditto for taking more time off work and having decreased coping. It just seems laughable to say these things give you fatigue, when the far simpler interpretation is that these are a consequence of the ongoing fatigue.
Snow Leopard
Senior Member (Voting Rights)
Yes, all of this needs to be emphasised again and again.
We should be hesitant to leap to conclusions about causality when the quality of the evidence is low.
Woolie
Senior Member
Ha, I just noticed: "female gender... may contribute to ongoing fatigue..."
We see what you did there. You just slipped that in, right next to the "psychological factors" without commenting on it directly. Its hard to believe that a team headed by a woman would stoop to such a tactic.
Or perhaps they are working up to recommending hormone treatment and gender reassignment surgery for all!
We see what you did there. You just slipped that in, right next to the "psychological factors" without commenting on it directly. Its hard to believe that a team headed by a woman would stoop to such a tactic.
Or perhaps they are working up to recommending hormone treatment and gender reassignment surgery for all!
Which, of course, still does not demonstrate a causal link, merely that it is not ruled out. A necessary but hardly sufficient step.
They also need to rule out the quite reasonable possibility that the apparent onset time point is not just the escalation of a pre-existing disease process from the prodromal (pre-clinical) phase to the full clinical expression phase. A much more difficult task, I grant, but no less necessary.
Is it reasonable to assume that disease processes suddenly switch on at point X in time, going from completely non-existent and having no discernable impact upon patients' lives, to a full or clear clinical presentation overnight (more or less)? It is reasonable to assume that disease has no significant burden until it explicitly presents in the clinic (or only for a short time before it presents)? No, it is not. There are no lack of counter-examples to that claim.
Far more likely is that, for many diseases at least, the underlying casual process(es) will have been at work for some time, maybe even decades in some cases, and causing significant impacts upon patients' general capacity long before diagnosis.
It is easy and lazy – and more than a little convenient for some – to interpret more generic non-specific symptoms often presenting during the pre-diagnosis phase (fatigue, pain, brain fog, etc), which are actually caused (directly or indirectly) by the undiagnosed disease, as non-disease factors playing a causal role in the disease. But it is wrong and unsafe.
e.g. Wessely's frequent assertion that a high level of depression/anxiety in ME patients must surely be a strong indication of primary psychopathology, instead of what it is much more likely to be, a completely understandable response to the unavoidable and causally opaque primary burdens placed on patients by the disease itself, in combination with the completely unnecessary secondary psycho-social burdens caused by irresponsible psycho-social speculation by the likes of Wessely.
There is no lack of examples of this problem in other diseases. It is a well known issue for neurological and endocrinological diseases, among others.
This is one the greatest dangers of the whole MUS project, that it will delay diagnosis and treatment for known and treatable diseases that have not yet reached full expression, with potentially fatal consequences in the worst case scenario, and significant additional unnecessary burdens upon the patient in the best case scenario. I have not seen any serious engagement on this problem by the MUS advocates, and it is one of their greatest technical and ethical failings. It is not enough to simply acknowledge it, to some degree, and then continue on as before. There needs to be a much more rigorous and explicit process for ruling these possibilities out, before the MUS label can be applied, assuming it even has any validity at all in the first place, a dubious proposition at this point.
I am not saying anything new. These kinds of issues have been presented repeatedly in the formal and informal literature for decades. Yet here we are, with them still not being properly taken into account.
Last edited:
Both causality and direction of causality.
Last edited:
Sly Saint
Senior Member (Voting Rights)
I've only just read this in competing interests
"TC is the author of several self-help books on chronic fatigue for which she has received royalties. TC (KCL) has received ad hoc payments for workshops carried out in long-term conditions. TC acknowledges financial support from NIHR. She has a patent background IP with a software company for which she receives fees for work unrelated to fatigue. There are no other relationships or activities that could have influenced submitted work."
as if that list isn't bad enough!
"TC is the author of several self-help books on chronic fatigue for which she has received royalties. TC (KCL) has received ad hoc payments for workshops carried out in long-term conditions. TC acknowledges financial support from NIHR. She has a patent background IP with a software company for which she receives fees for work unrelated to fatigue. There are no other relationships or activities that could have influenced submitted work."
as if that list isn't bad enough!