Creating space:

This isn't really relevant to the CFS article, so I thought I'd discuss it here. I don't really follow your thread. First of all, there are multiple studies using different methodology (Canadian census + CFS patients and matched controls), and the matched control study has been replicated by the authors. The results show that CFS patients do have higher reported incidence of child abuse than non-patients, and it seems pretty robust. It also fits in with the most likely etiology of CFS, i.e. dysregulated stress system. --sciencewatcher (talk) 19:56, 13 August 2011 (UTC)[reply]

Heim et al 2009 is a replication of Heim et al 2006 but by the same authors and employed the same problematic methodology, eg it was retrospective and as previously explained used flawed CFS criteria. I also had a closer look at the Canadian consensus census study [1] - it focuses on childhood physical abuse, they did not ask about sexual abuse or emotional abuse and neglect, they did ask about other psychosocial factors but used them to adjust/control the association with childhood physical abuse. It is good to have another large study, but unfortunately it suffers from the same types of problems discussed previously, most of which the authors mention under "Limitations" on p461, and I wish they had included control diseases.

The researchers merely asked individuals if they had been diagnosed with CFS (etc) by a "health professional" which doesn't instill much confidence in the accuracy of the diagnosis, it is not clear who qualifies as such. Diagnosis of CFS is depends heavily on subjective clinical judgements and is open to opinions subjected to bias. The CDC claims that 80% of CFS cases are undiagnosed [2] and the results from Newton et al 2010 [3] and the PACE trial [4] would suggest that at least a third of people being referred to or from CFS clinics in the UK, cursorily assumed by physicians to probably have CFS, don't actually meet CFS criteria when examined properly.

So the Canadian consensus census study is evidence that retrospective self-reported childhood physical abuse is associated with a small increased risk (OR = 2.11; 95% CI = 1.22, 3.65) of self-reported CFS diagnosis made by a health professional. Nothing more and nothing less. Once again, no actual evaluations and essentially "CFS-like". From memory, the CFS-like classification from Kato et al 2006 and Huibers et al 2004 discussed earlier at least asked about CDC 1994 criteria symptoms.

For CFS(like), the Canadian consensus study found a weighted prevalence of 1.3% (95% CI = 0.8, 1.8). Similarly, note that in a large CDC study (see Table 4), 1.607% of the population sample were classified as "CFS-like", however only 0.235% were diagnosed with CFS after proper assessment, out of the CFS-like individuals only about 1/7 met CFS criteria.[5] (Out of the 1155 who reported CFS symptoms over the telephone, 52% also reported a medical/psychiatric exclusion when asked. Of the remaining 555 without such exclusions but now "CFS-like", 46% declined further assessment; of the remaining 299 that underwent a clinical examination, a further 47% had a medical/psychiatric exclusion, 37% had insufficient symptoms, and only 14% had CFS.) All this makes the CFS-like classification almost meaningless, and studies which use it may be automatically suspect.

SW, surely you're also aware that memories of childhood are not necessarily or particularly accurate and recall can be biased by a number of factors. While not synonymous with CFS, it is also often assumed that childhood abuse is a major risk for medically unexplained chronic pain. PMID 14980144 is a review which reports that retrospective studies support the association but prospective studies refute it. PMID 11323150 was both a prospective and retrospective study which found this phenomenon of discrepancy occurred in the same group of people. PMID 21050659 suggests PTSD mediates this effect. That was just a quick sample and the point is not that such associations aren't possible (I'm sure there are some prospective studies out there supporting a link between childhood psychological factors and adult diseases, for example arthritis, PMID 19251363 and PMID 15490926 with effect sizes that don't support psychobabble). The point is that care should be taken to search for actual robust evidence before placing such confidence in a supposed association based on one or two retrospective and problematic studies and then repeatedly calling the evidence "pretty robust".

A stress induced HPA-axis related neurodevelopmental disorder as postulated by Heim is an interesting model in general, but it is a major red flag that as soon as the CDC changes the way it diagnoses CFS, which seems to further confound it with major depression and probably other psychiatric diagnoses, suddenly the prevalence estimates go up 10 fold, and contrary to limited previous research childhood trauma suddenly shows up as a risk major factor with effect sizes higher than what is usually seen in classic mental disorders. The "most likely etiology of CFS" is a matter of opinion, and a "dysregulated stress system" not only needs etiological context before making psychosocial speculations based on limited evidence but is also an inappropriate generalization if based largely on HPA axis findings which are rather mild and aren't even found in the majority of CFS patients. - Tekaphor (TALK) 01:29, 27 August 2011 (UTC)[reply]

I agree with Tekaphor's analysis (though note typographic confusion of Candian Census with Canadian Consensus). The proposition of a single aetiology in a Syndrome characterised only by symptomology is of dubious logic and the extrapolation of the results from a small study which represents a narrow cultural range (US population), to a patient group that is of reported global spread, requires far more robust support than the hypothesis of "dysregulated stress system" currently provides. The Heim et al study produces improbably high figures for multiply experienced sexual, emotional and physical absuse (35 of 106) which if translated to the 2.5% prevalence for CFS which Heim uses (again this is an improbable number and seems based on the flawed Reeves work - questioanaire based, no physician examination) - would suggest that 8% of all sexual abuse victims in the US would have CFS [[6], something which should be evident in abuse prevalence studies. It is also notable the study controls appear to be unrepresentative in that they are underreporting levels of abuse by a factor of 2 compared to population studies [[7]]. Heim et al may well have identified an important phenomenon ("dysregulated stress system") but the relationship of that phenomenon to CFS requires studies amongst other populations experiencing other types of stress in childhood, to answer questions such as "does sexual abuse have a greater impact on disregualtion, than starvation or exposure to military conflict ?" If the stress hypothesis were to be corect,starvation and conflict would seem likely to produce a dysregulatory effect, with consequent CFS symptomology, given the volume of longtitudinal studies carried on War II childhood survivor populations, it's surprising that no association with CFS or CFS comparable illness has so far been identified. The lack of any notable presence of CFS in the Dutch Famine study might be considere significant given that prenatal ACE's are noted of of particularly significance to the prevalence of other illnesses develoing in adulthood [[8]]

The Canadian Census study suffers from three major limitations, firstly there is no moderation of the self report of CFS diagnosis - an original diagnosis such as PVFS, or myalgia may have been given many years previously, to which either the patient or a non specialist health worker has ascribed the term CFS, meaning that there was no certainty of diagnostic consistency across the study. Secondly, as with Heim et al, there was no verification of the report of ACEs, and thirdly, the survey area is of a very particular geographic and demographic context - the Canadian provinces of Manitoba and Saskatchewan. The study results are interesting, but are not open to meaningful extrapolation.

--In Vitro Infidelium (talk) 16:32, 29 August 2011 (UTC)[reply]

Looking at the overall science, you have to admit that stress is the most likely hypothesis...most CFS patients say their illness was triggered by a stressful period in their lives (and this shows up both on informal questionnaires posted on websites, as well as in proper scientific studies). Also the various signs (HPA axis dysregulation, immune/ANS dysregulation, abnormal blood volume, SPECT scans) and symptoms (fatigue, brain fog, depression, anxiety, pain, dizzyness, etc) are all typical of a stress-related illness. The only argument I ever hear against stress is that many people don't like it - they think it means their illness was either imaginary or 'not serious' or 'my own fault' or something similar. Basically the argument against the stress hypothesis is based on emotion and not science as far as I can see. Also I see on your website that you say you 'believe' that CFS is organic (i.e. not caused by stress), so I don't really see any point continuing this discussion if you have already made your (closed) mind up. I think Tekaphor has a similar belief that CFS "must" be organic, although he/she hasn't explicitly said that. --sciencewatcher (talk) 18:03, 29 August 2011 (UTC)[reply]

I have no idea what a "most likely hypothesis" is, all hypotheses fulfulling the basic standards of logic are equally valid until excluded by falsification, probability has nothing to do with it. Given your rapid recourse to ad hominem responses, you appear to operating under a false pseudonym as neither your reasoning nor your conduct accords with the valuing of scientific principles. What I believe about M.E/CFS has nothing to do with Wikipedia, however what I have written at: http://cfsmirror.blogspot.com/ is "from a perspective that acknowledges that most people with a physician given CFS diagnosis, have a disabling organic illness." For what it's worth (it is after all merely my personal perception so in scientific terms, not much) I'm quite prepared to accept that M.E/CFS is a disease description that covers in some measure within individuals and across the patient population, elements of somatisation processes that are either contributory or even causative. At this stage however there is no compelling evidence to suggest single causation across the patient group. Hyping the 'stress hypothesis' appears to me to be no more valid than 'hyping the XMRV (and its inevitable offspring MULV, HGRV etc) hypothesis. We can all accuse other people of having dubious motivations for holding the opinions that they do, and label them as having closed minds because they disagree with our own perceptions of what is true or valid or 'proper science' - but that is neither scientific, nor is it condusive to a collegiate process of creating an encyclopaedia. --In Vitro Infidelium (talk) 13:34, 30 August 2011 (UTC)[reply]

Well you have to agree that a large number of CFS 'advocates' are hugely biased and profess the ignorant and incorrect viewpoints I pointed out above (and you yourself point this out in your blog). If you're having a scientific discussion it is important to find out whether the other person has dogmatic beliefs - if so, you are just wasting your time. I know someone who works at a museum and they get a lot of creationists. They try not to get into a scientific discussion, because it's a waste of time. I got the ironic impression that you yourself also had this biased viewpoint based on the comment on your blog, but perhaps I got the wrong end of the stick. It's not bad to point out the obvious, and I have to do it quite a lot on wikipedia when people push crap science because they "believe" that MSG or HFCS is dangerous or that low-level EMF from cellphone towers will give you headaches or whatever (check my recent edits).

Getting back to the discussion: the stress hypothesis seems to be the most likely because it has the most compelling evidence supporting it (unlike XMRV, which has been pretty much disproven). Hypotheses are not all 'equally likely' if you have evidence supporting some of them and rejecting others. Although as tekaphor points out above there is a lot of variability in HPA axis studies on CFS patients, there does seem to be some dysregulation although it is hard to pin down exactly (similar for the immune system). And contrary to your comments above, CFS-like illness is seen in war, in gulf war syndrome (and similar combat syndromes in all previous wars). --sciencewatcher (talk) 15:09, 30 August 2011 (UTC)[reply]

Also, I just checked out that dutch famine study. None of the links work in firefox. It's a very strange website - all the links are actually images that don't have any anchors! Anyway, from the summary text it says that 'We have found no evidence for HPA programming' which would seem to suggest it would be unlikely to cause CFS. Also, did they actually look for CFS? If not then they aren't likely to find it.

Stress is the only hypothesis that seems to fit all aspects of the illness. In my own non-professional research of the illness, stress does seem to fit the entire patient population. That view is from both looking at the science and talking to hundreds of patients. I don't have any particular beliefs or prejudices about the illness. I'm open to any explanation. But so far the only one that seems to fit all aspects of the illness seems to be a dysregulated stress system. This is also the theory that seems to be held by the CDC. If you look at the pathophysiology page here, all the main theories that have any significant evidence are behavioural/stress (which is essentially the same thing anyway). What evidence is there that any other hypothesis is more valid, or that stress doesn't fit any particular group of patients? --sciencewatcher (talk) 17:12, 30 August 2011 (UTC)[reply]

I fully understand the difficulties of trying to maintain logical integrity within a discussion about matters which are expressed in scientific terms, with those who are either opposed to the very rationale of scientific enquiry or, and even more problematically, those who have adopted a ‘partialist’ attitude – the ‘pick ‘n’ mix approach of choosing evidence which fits, and rejecting that evidence which doesn’t. However context is everything and not all circumstances demand that scientific absolutes are the defining aspect of discourse, so on the IVI Blog (a stalled project) I declare the perspective from which it is written. In contrast when participating on WP I’m happy to work from a perspective of ‘encylopaediasm’.

As I wrote previously, I don’t understand how post Popper, one can arrive at a reasoned probalistic ordering of hypotheses. A preponderance of evidence argument can only apply when all conceivable null hypotheses have been tested; absence of evidence does not and can not imply evidence of absence. There have been probalistic ‘meta hypothesis’ proposals suggested as way of identifiying the most rewarding route for a give research direction, but that isn’t the same thing as ariving at a definitive statement of ‘most likely correct. The notion of ‘fit’ seems to me to be a non scientific line of argument, more forensic than scientific although I suppose it is nosologically valid, though I would take that as a criticism of nosology rather than as support for a ‘good fit’ approach .

As to the validity of the ‘stress hypothesis’, at face value it appears scientifically valid however I’m not sure it is obviously falsifiable, there are too many potential confounding factors which your own arguments seem to embrace. The strength of using a study of ill adults who report ‘childhood trauma’ is that it implicates a developmental process: ‘insult to the developing organism produces chronic effect exressed in adulthood’. However such an implication has to be tested to have validity in supporting an hypothesis, correlation does not equal cause and it is necessary to resolve how and why trauma in various forms, impacts upon developmental processes.

To favour an hypothesis which relies on reports of sexual abuse, without any comparison to other traumatic insults – starvation, disease etc is not justified. The Dutch study is very useful in that it allows a rare comparison of a group of fetally nutrient deprived people with close contemporaries who enjoyed no such deprivation, against a background of otherwise stable cultural and economic conditions. How much work the Dutch researchers have done on fatigue I don’t know, but for anyone wanting to explore the ‘stress hypothesis’ the Dutch study offers a wealth of relevant data. To invoke adult experience of wartime trauma as comparitive to ‘childhood trauma’ is logically fallacious because it removes the issue of ‘developmental insult’. Adult experience of trauma as a supporting characteristic of the ‘stress hypothesis’ incurrs multiple confounding factors, which can only be addressed if there is physical data both pre and post the trauma experience. Without such data one is left with a circularity: Is the ‘poor’ response to ‘trauma’ a function of pre existing physiology, or is the post trauma physiology a response to a pre existing pyschological debility.

An hypothesis trading argument is futile, and IMO unscientific. M.E/CFS is defined purely by symptomology, the symptoms are multiple and variable across a large population – probabalistically it would be highly remarkabe if a single disease process were to underly the ill health of the globally affected population. There may well be commonalities across the population – one or more genetic predispositions, fetal or infant developmental insult that may include disease, nutritional impairment, evironmental exposure or emotional stress; disease, nutritional impairment, evironmental exposure or emotional stress having developmental impact in adolescence or, disease, nutritional impairment, evironmental exposure or emotional stress experienced in adulthood. From such reasoning one could postulate that the only valid hypothesis is one that allows that: “M.E/CFS is a condition that results from multiple impacts, with no single causative combination of impacts present throughout the global population”.

--In Vitro Infidelium (talk) 12:18, 6 September 2011 (UTC)[reply]

Thanks IVI for pointing out the embarrassing typos which confused census for consensus, I have corrected them with strikeout. Sciencewatcher, you claim that the "stress hypothesis" (the flavour where psychological stress primarily causes and perpetuates only functional biological abnormalities resulting in CFS which are automatically corrected upon removal of the alleged psychological stress?) seems to "fit all aspects of the illness", "fit the entire patient population", you are "not aware of any evidence that it doesn't". I addressed the research on psychological stress and HPA axis function on the main CFS talkpage [9]. Even if we ignored the studies which found no such association with preonset stress and assumed the remaining limited evidence for the stress hypothesis wasn't problematic (including the vague biological correlates), it still doesn't apply to roughly half of patients or about 1/3 - 2/3 depending on the issue or study. The hypothesis that severe chronic stress can induce or contribute to CFS-like symptoms is not implausible; what is illogical however is the blanket application to all patients by default within the heterogeneous wastebasket such as broadly defined CFS. Therefore, you can no longer seriously claim to be unaware of any evidence that challenges this. Persisting with such a blatant disregard for the conceptual and methodological issues involved would be cringworthy. Further presenting oneself as a vanguard against bad science while displaying a massive blindspot for mind over body ideology would also be an embarrassment to rational skepticism, especially when portraying patients and advocates as dogmatic with huge biases and incorrect viewpoints but oneself as unbiased and objective. I will comment on the rest another time, still working on a response. - Tekaphor (TALK) 02:58, 5 October 2011 (UTC)[reply]

Maybe this off-topic conversation has gone cold and these words will fall on deaf ears, but here is the rest of my comments on claims made above by Sciencewatcher:

• The claim that CFS symptoms (selectively chosen to support the stress hypothesis) "are all typical of a stress-related illness": This is misleading because CFS symptoms are similar to such a wide range of conditions that CFS is fundamentally a diagnosis of exclusion. In the CDC study presented earlier (Reyes et al 2003) medical exclusion of people with CFS-like symptoms was very common (much more common than psychiatric exclusion), including diseases not thought to be "stress-related" illnesses. A similar problem exists for the statement "(HPA axis dysregulation, immune/ANS dysregulation, abnormal blood volume, SPECT scans) are all typical of a stress-related illness". None of those abnormalities apply to all people meeting CFS criteria or any other so-called stress-related illness, and can occur in organic disease too. Imprecise and vague similarities that only occur in some patients are not conclusively indicative of anything and could be secondary or indicate subtypes.

• The claim that the stress hypothesis has "the most compelling evidence supporting it": This isn't saying much for CFS research in general, you have to admit that CFS research in general is poor. Your bias is obvious when you seem to generally think of the biomedical evidence as all pretty dodgy but the psychological evidence as all "pretty robust" despite demonstrable flaws. Rejecting potential biomedical and biological factors because they aren't universal or necessarily explain etiology would be ironic because the same applies to all the psychological factors you apparently lump together. You claim that the "only" argument against it is that "many people don't like it" for downplaying the condition, but this claim is disingenuous because the relevant research itself and the weaknesses contained therein are a major argument against blanket application, as mentioned above and discussed on the CFS talkpage [10]. Conversely, at times it has seemed that you "don't like it" when issues with psychological research and associated hypotheses are raised for questioning?

• The claim about patient anecdotes and alleged biases: Talking to X number of patients whose experiences supposedly support your views anecdotally doesn't mean much when there seems to be a similar number of patients whose experiences contradict it. Perhaps your own personal experience of CFS as a psychosomatic illness caused by stress and cured by changes in thoughts and behaviour is naturally influencing your perception in the same manner as other patients who had the opposite experience and therefore became skeptical towards the role of cognitive behavioural factors? You claim to have no "particular beliefs" regarding CFS and talk about the importance of finding out whether the other person has dogmatic beliefs before engaging in scientific discussion. However, your perpetual insistence that the stress hypothesis applies to all patients, in clear violation of the evidence, is a sign of a dogmatic belief, so something isn't adding up here.

Asking people with a suspected "dysregulated stress system" whether they felt "stressed" years ago may be as unreliable as asking chronic pain and PTSD patients whether they experienced childhood abuse, as per previous conversations. A significant proportion of patients of some organic diseases, eg breast cancer, also retrospectively report or "believe" that stress was significant at the onset of their condition but this doesn't mean it is true (the prospective evidence is somewhat conflicted but leaning towards "untrue"). On the other hand, stress has been reportedly associated (retrospectively and prospectively) with the onset and progression of a wide range of diseases, including ones not usually viewed as "stress-related illnesses", but again without effect sizes that justify psychobabble and/or primary stress hypotheses. There are some diseases where mental symptoms may be among the first to manifest before physical symptoms, and therefore encourage false attribution of the latter to the former. There are also some non-CFS studies which question the role of stress in idiopathic prolonged fatigue too (which you seem to think is the same) and suggest stress and fatigue are independent.

The above hasn't considered the biological evidence, or issues with ME/CFS case definition, or detected biological and psychological differences between CFS vs GWS vs FM vs IBS. Yes, many of the biological findings, without etiological context, may reflect either organic or functional or psychological abnormalities further upstream and until etiology is better established this debate is going to continue ad nauseum because whatever is found will be attributed to either mental or physical causes depending on ones own understanding and/or confirmation bias. However, the prevalence and levels of psychological stress and supposedly related HPA-axis abnormalities involved do not explain the severity of physical illness that follows. Relying on stealth psychological stressors and Freudian-style denial and lack of insight to routinely explain failed recoveries of patients after external/internal stress reduction and why large proportions of patients don't report such stressors, is unsound and currently unfalsifiable so something else is probably going on as well or instead.

"Stress" is ambiguous and somewhat ubiquitous. The vast majority of people in the general population under severe stress do not develop CFS, even the ones under chronic internal stresses. Frankly I think you have taken one plausible contributory factor and blown a lot of hot air into it, which seems common in CFS speculations, including biological based ones. It is much more difficult to discover and understand novel biological mechanisms for complex diseases than it is to make premature presumptions about vague psychosomatic mechanisms or simplistic appeals to psychoneuroimmunology. It just seems a little more complicated than that. Psychological medicine or equivalent fields have a long history of sloppily overstating psychological factors and misattributing symptoms as functional. This of course doesn't mean the psyche plays no role or that symptoms cannot be functional, but illness usually turns out more complex than people initially realise even when stress is involved, and it would be hubris to believe that in the early 21st century we have reached a point where we cannot repeat the errors of previous generations in modern ways.

I don't think CFS "has" to be organic in a classic sense, it seems to be a wastebasket heterogeneous syndrome anyway, and I don't think there is a really strict division between organic and functional. I also reject the notion that "functional" necessarily equals "psychosomatic" in the sense of being reversible with cognitive behavioural interventions. I don't think a "bodily distress disorder" and biomedical factors are mutually exclusive either and I believe the grey area is more subtle and biological than most psychosomatic ideologues realise. IMHO among CFS is a condition more rooted in biology than psychology, regardless whether mental factors are significant at some stage.

I don't have a problem with stress contributing/perpetuating CFS, even to an organic ME if it exists, or the involvement of a dysregulated stress response system with as yet unknown etiology. There needs to be better conducted research on the issue before the association can be confirmed or the nature of this association can be explored. But I do have a problem with sweeping generalisations based on distortions of flawed evidence and premature conclusions, especially when coming from people presenting themselves as practitioners of skepticism against poor science and CAM in general but then displaying massive oversight when it comes to cognitive behavioural factors in CFS. - Tekaphor (TALK) 03:48, 4 November 2011 (UTC)[reply]

Hi! I wanted to let you know I moved your page that you created in the article space to User:In Vitro Infidelium:You/CFS Parsimony. It looked like maybe you just forgot the "User:" prefix. Cheers! --Nick—^Contact/_Contribs 17:39, 9 March 2015 (UTC)[reply]

And now I've done the same with User:In Vitro Infidelium/CFS Parsimony. --Nick—^Contact/_Contribs 17:43, 9 March 2015 (UTC)[reply]