Discussion in 'PsychoSocial ME/CFS Research' started by Tom Kindlon, Aug 15, 2019.
redefinition as a career promoting move....?
8.6% of the general population? That is absurd.
Wow, this is more than 20 x times of what credible epidemiological studies have found. And they don't even explain how far off their definition is from diagnostic criteria:
I think that's the highest prevalence estimate I've ever seen.
Therapists will be pleased they can now potentially give CBT/GET to many more people, or at least that seems to be the goal here.
Am I reading this correctly? 8,6% of the population sample said they had fatigue symptoms more than usual compared to last month, and the authors labelled them as CFS patients?
The virtues of Per-reviewed research. I mean it's hard to look at it any other way...
bodily distress syndrome (BDS).......
except this is not a recognised 'disorder' (apart from Per Fink)
Can anyone just make up a name for an illness and publish research on it?
(am thinking of Esther Crawleys CDF as well)
Well, the Danish have convinced the world they have been producing salami for several decades now. Surely this is much the same post-truth Deleuzian, best back, there's more where this came from, handle-turning? The only question is when does it disappear up its own jejunal villi.
But Per-reviewed research - that is a touch of genius.
How are those clowns taken seriously? I'm surprised they left TBI and migraines off the list.
Not even bothered with properly diagnosing. Just questionnaires.
The cluelessness on display is simply galactic-sized. Absolutely nothing of value can be concluded from such a study.
As otherwise the main takeaway is that according to assumptions made by this kind of researchers, there is currently a massive epidemic of disabling psychogenic illness that affects roughly 1/12 human. Of unspecified origin, or mechanism, or explanation. As long as you keep everything as vague as possible and mix-and-match according to an arbitrary set of beliefs.
It would represent the single largest "disease" in the world, likely affecting close to 10% of the world's economic output. No one describe its mechanism, or anything even remotely related to an explanation. It just exists and the solution is therapy. Because reasons.
Ridiculous. A complete waste of everything.
Clearly they have a rather different definition of CFS.
From the participants and methods section:
From the discussion section:
I quickly skimmed the referenced study . It seemed a bit muddled and talks mainly about 'chronic fatigue' and some 'CFS-like illness' but at least, unlike the present study, it recognises the limitation that "we did not use an interviewer-based detailed definition of chronic fatigue preventing us from extrapolating our findings to this smaller group of the more severe chronic fatigue syndrome."
They almost had an insight here, but then they go and spoil it again:
There is supplementary material where they explain how CFS was defined. See: https://journals.sagepub.com/doi/suppl/10.1177/1403494819868592
They only used the Chalder Fatigue Questionnaire as expected but with one difference: they asked about fatigue symptoms in the last 12 months. The CFQ usually asks about symptoms in the last month or when participants were last well. I would be clearer if they had added precisely what the participants were asked- how the questions were formulated.
They also report in the paper that "A subgroup analysis was made excluding participants who reported suffering from chronic physical disease which could possibly account for their symptom patterns." They reported this in the supplementary material, in Tables S10–S14 and Figures S2–S3. I suspect there is a mistake in the title of Table S10 and that these are the prevalence rates adjusted for chronic physical disease in both men and women, not just women (which is reported in Table S11).
If that's the case, it's rather strange how little difference, adjustment for chronic physical disease makes to the prevalence rates of the 'syndromes'.
Yes, they say that other studies have reported higher prevalence estimates than their study, but that doesn't seem right for CFS. Reference 15 is about unexplained chronic fatigue, which the authors acknowledge is different from CFS: "It is worth noting that chronic fatigue is much more common and relevant to primary care, than chronic fatigue syndrome."
Peterson et al. also write:
Reference 14 refers to the LifeLines study by Judith Rosmalen. The figure of 11% they report is of a self-reported lifetime diagnosis of CFS, FM or IBS. But look at the distribution of that 11%: it's very different from what Peterson et al. reported (the number don't add up to 11% because of overlap)
Even if you don't know anything about CFS and how it is diagnosed, if you have read research on FSS it should be obvious that these numbers of the DanFunD do not seem correct. The prevalence of CFS is higher than that of FM and IBS combined!
It gets even weirder. Apparently, DanFunD does have a lot of information about its participants. You can read about it here: https://pdfs.semanticscholar.org/7e...1.1480774417.1566033579-2114742670.1563967729
Each participant had a clinical examination. The researchers have blood samples, fitness test results, pain pressure thresholds and much more:
If all participants were clinically examined into detail and some received a telephone-based diagnostic interview, it should be possible to come up with a more realistic definition of CFS than scoring more than 4/11 on the Chalder Fatigue Scale.
In fact, they have data on the self-reported prevalence of the functional somatic syndromes. It's approximately 1,19% for CFS, much lower than the 8,6% reported in their recent study. So they could have either used more information about the participants to come up with a realistic CFS definition or - if that was too difficult - they could have used the 1,19% self-reported CFS.
But instead, they prefer the 8,6% estimate (that's the estimate where even chronic physical diseases that account for fatigue symptoms are not excluded for some reason) which totals 823 CFS patients. I've found some conference abstracts in the Journal of Psychosomatic Research, where Fink's research team seems to use that broad definition of CFS to perform analyses on the valuable data they have on participants of DanFunD.
There's one on lifestyle parameters, which report n = 823 CFS patients, more than the number of FM and IBS patients combined.
One on lipid- and glucose metabolism which also reports n = 823 CFS patients, more than the number of FM and IBS patients combined.
And one on conditioned pain modulation and pain sensitivity. Here the n for CFS is 187, but it's still more than the number of FM and IBS patients combined. So presumable it's just a subgroup of patients using the same broad definition for CFS.
It would be unfortunate if all this valuable information that was gathered on a representative sample of the Danish population was used to do analyses on a CFS definition that doesn't make much sense.
One could argue that they are simply studying chronic fatigue and just made a mistake in naming it. But that will confuse the literature, much like studies of the empirical criteria (the Reeves definition) have done. I suspect that in future publications that use the broad n = 823 CFS sample of the DanfunD for analyses, it will no longer be clear to readers how this sample was formed, that the prevalence of CFS was 8,6% etc.
Another problem is that, in my view, CFQ scores do not form a good definition of people suffering from chronic fatigue either. They only had some fatigue symptoms more than usual within the last 12 months. Perhaps people were more fatigued because they were renovating their house or just had their first baby. With the CFQ, we don't know anything about the intensity or frequency of fatigue or whether it could be explained by exertion or stressful life events.
I am reminded of a comment at the other place (a long time ago) from @Lucibee
"the only tool a statistician should use on the CFQ is a sharp stick"
And like all the CFS studies that used Oxford and quite a few with Fukuda. In a review of 53 Fukuda studies, Jason reported that the occurrence of PEM ranged from 25-100 percent of patients in those studies while unrefreshing sleep ranged from 16-100 percent. Both symptoms are required by CCC and IOM for a case of ME and significant in ME-ICC. Its impossible to know what these CFS cohorts are and it just muddles the picture with findings that are best suspect and at worst, erroneous and irrelevant. But they nonetheless can end up in "evidence-based" clinical recommendations
The logical conclusion is that a score of 4 or more on the Chalder Fatigue Scale is not a specific predictor of CFS. (Note that Chalder reported the cutoff score of 3 or less as indicative of "normal" fatigue using bimodal scoring, not Likert scoring.)
Anyone who has actually looked at the questionnaire knows how incredibly nonspecific this cut off would be.
Maybe worth noting that the Chalder Fatigue Scale was not included on the list of fatigue instruments recommended as part of NIH's Common Data Elements for ME/CFS.
Explanation of terms... the Core designation is for instruments expected to be used in all studies. The exploratory label is when e.g. not previously used in ME/CFS, needs further validation, etc.
Separate names with a comma.