Talk:Myalgic encephalomyelitis/chronic fatigue syndrome/Archive 19

History section rewritten using IoM
''The IoM report offers more than just a good source, it is an extensive piece of research about the whole context of CFS. In the case of the history of the disease the IoM helps close a number of gaps and inaccuracies, I've taken the quoted sources in most cases rather than the IoM itself to produce a new rendering of the history section which I'm posting here for comment. I've not included the current paragraph on XMRV, which personally I'd prefer to see go, but previously it was felt this should stay, my proposed text would not affect this. There are two references which editors should give particular consideration to, one is to the IoM itself re: Ramsay and the other is to Speight 2013 where the publication might not be thought adequate - but my thinking is that if the IoM has quoted it then validity is established.''

Myalgic Encephalomyelitis

From 1934 onwards, outbreaks of a previously unknown illness began to be recorded by doctors. Initially considered to be occurrences of  poliomyelitis the illness was subsequently referred to as “epidemic neuromyasthenia”. In the 1950s, the term “benign myalgic encephalomyelitis” was used in relation to a comparable outbreak at the Royal Free Hospital in London. The descriptions of each outbreak were varied but included symptoms of malaise, tender lymph nodes, sore throat, pain, and signs of encephalomyelitis. The cause of the condition was not identified although it appeared to be infectious, and the term “benign myalgic encephalomyelitis” was chosen to reflect the lack of mortality, the severe muscular pains, evidence of damage to the nervous system, and to the presumed inflammatory nature of the disorder. . The syndrome appeared in sporadic as well as epidemic cases and in 1969, benign myalgic encephalomyelitis appeared as an entry to the International Classification of Diseases under Diseases of the nervous system.

The authors of a review of 15 outbreaks of benign myalgic encephalomyelitis carried out in 1970, concluded that these were psychosocial phenomena caused by either mass hysteria on the part of the patients or altered medical perception of the community. . These conclusions were based on the higher prevalence of the disease in females in whom there was a lack of physical signs,on that basis the authors recommended that the disease be renamed “myalgia nervosa.” Despite strong refutation by Dr. Melvin Ramsay, the proposed psychological etiology created great controversy and convinced health professionals that this was a plausible explanation for the condition.

The continued work of Dr. Ramsay demonstrated that, although the disease rarely resulted in mortality it was often severely disabling", because of this Ramsay proposed that the prefix “benign” be dropped   . In 1986, Dr. Ramsay published the first diagnostic criteria for ME, in which the condition was characterized by:

•	a form of muscle fatigability that even after minimal physical effort, 3 or more days elapse before full muscle power is restored.

•	extraordinary variability or fluctuation of symptoms even in the course of one day.

•	an alarming chronicity.

Chronic Fatigue Syndrome

In the mid-1980s two large outbreaks of an illness which resembled mononucleosis drew national attention in the United States. Located in Nevada and New York, the outbreaks involved an illness which was characterized by “chronic or recurrent debilitating fatigue and various combinations of other symptoms, including sore throat, lymph node pain and tenderness, headache, myalgia, and arthralgias”. An initial link to the Epstein-Barr virus saw the illness acquire the name “chronic Epstein-Barr virus syndrome”.

The United States Centers for Disease Control and Prevention convened a working group tasked with reaching a consensus on the clinical features of the illness. Meeting in 1987, the working group concluded that CFS was not new and that the many different names given to it previously reflected widely differing concepts of the illness’s etiology and epidemiology. . The CDC working group chose “chronic fatigue syndrome” as a more neutral and inclusive name for the illness but noted that “myalgic encephalomyelitis” was widely accepted in other parts of the world. . The first definition of CFS was published in 1988 and although the cause of the illness remained unknown there were several attempts to update this definition, most notably in 1994,. In 2006, the CDC commenced a national program to educate the American public and health care professionals about CFS. ''

--In Vitro Infidelium (talk) 10:33, 3 December 2015 (UTC)


 * Is there a reference for "In the mid-1980s two large outbreaks of an illness which resembled mononucleosis"? From what I understand, the link with mononucleosis and thus EBV had more to do with the recent availability of EBV serology testing and the finding that some patients had elevated titers. The symptoms I've seen discussed associated with that outbreak have little to do with symptoms of mononucleosis. Anal0gue (talk) 00:00, 7 December 2015 (UTC)


 * The three sentences are referenced to Holmes 1988 - see the ref list number 15. The relevant IOM text = In the mid-1980s, two large outbreaks of an illness in Nevada and New York resembling mononucleosis attracted national attention. The illness was characterized by “chronic or recurrent debilitating fatigue and various combinations of other symptoms, including sore throat, lymph node pain and tenderness, headache, myalgia, and arthralgias”(Holmes et al., 1988, p. 387). The illness was initially linked to Epstein-Barr virus and became known as “chronic Epstein-Barr virus syndrome”(Holmes et al., 1988). The section under discussion is "History" - the issue is not what is medically correct, but what at the relevant time was believed or hypothesised or interpreted as being correct. --In Vitro Infidelium (talk) 17:21, 7 December 2015 (UTC)


 * As there have been no objections I've go ahead and done the full swap, plus added an IoM ref alongside the Holmes 1988, for support, the IOM being easily accessible.  I have also moved the the paragraph on XMRV to the controversies section where it more properly belongs, the move has the advantage of leaving the history section as a clear narrative of what has led to the present. XMRV had no impact on management, diagnosis etc, XMRV was always controversial and disputed, and never received any significant level of expert support.--In Vitro Infidelium (talk) 14:05, 10 December 2015 (UTC)

CFS subarticles
The CFS article has an exceptional number of sub articles, a total of 6. In comparison neither Fibromyalgia nor Rheumatoid Arthritis have any subarticles. The Multiple sclerosis article does have five subarticles but this is an illness for which there is a substantially greater body of MEDRS. According to http://stats.grok.se/ the main CFS page received 30267 visits in November 2015, the subarticles with their November visitor numbers are:


 * Pathophysiology of chronic fatigue syndrome 438


 * Clinical descriptions of chronic fatigue syndrome 384


 * Chronic fatigue syndrome treatment   1018


 * History of chronic fatigue syndrome 495


 * Alternative names for chronic fatigue syndrome 599
 * Controversies related to chronic fatigue syndrome 853

Maintaining these pages, particularly keeping them in line with updates to the main page involves effort and it is clear that several have been barely touched in years. The reason why there are so many sub pages seems perhaps more to do with the history of edit wars than actual encyclopaedic need. Two sub articles are in reasonable shape: Clinical Descriptions and Treatment, and both of these can probably be fairly easily improved with material from the IoM and P2P reports. The other four seem to me to be in a fairly parlous state, suffering from multiple problems including in some parts substantial WP:SYNTH. My thinking is that these are candidates for redirection back to the main article as per WP:RDR which would leave the option that the articles could be rebuilt at some future date if the need is established. With the four articles effectively deleted, this would allow concentration on improving and maintaining the main CFS article and the two subarticles in a cohesive manner. --In Vitro Infidelium (talk) 18:04, 12 December 2015 (UTC)


 * Yes, I would tend to agree. The original reason for creating them was, I believe, to cut down the length of the main article. However a lot dubious speculative stuff has already been removed from those sub-articles, and at the moment they aren't actually very long. I think it shouldn't take too much effort to merge them in. --sciencewatcher (talk) 23:01, 12 December 2015 (UTC)


 * The pathophysiology sub article was written by people who didn't understand what pathophysiology is. They put in a lot of vague speculation about causes. This isn't pathophysiology. A good pathophysiology article should list objectively measurable abnormalities that are commonly found in patients. Abmayer (talk) 22:48, 14 December 2015 (UTC)


 * I don't think it's fair to criticise the understanding of editors who may have had far fewer MEDRS to work from, who had numerous apparently contradictory lesser sources to address and who had to work in an atmosphere in which edit warring often errupted. That said I agree with the critique of the content as it currently stands, and without additional MEDRS I don't see the current pathophysiology section of the main article being capable of improvement or expansion to such a degree that a sub article would be required. --In Vitro Infidelium (talk) 17:05, 15 December 2015 (UTC)

Redirect proposal for Alternate names subarticle
Proposal to redirect Alternative names for chronic fatigue syndrome article – as discussion above. The sub article is a mish mash of material which lacks encyclopaedic value. The article as written is problematic because it presents what are in effect differing conceptions and then presents them as reducible to differing names for a single disease.

There was some material on the sub article page that was missing from the main article – I’ve corrected this with edits to the History and Controversies section. Past editing has resulted in a non encyclopaedic conflation of historical naming with the ongoing and current ME-CFS (and latterly SEID) debate in clinical contexts, I've put what are now largely redundant names in the history section. Conflation of media controversy with medical controversy is also unencyclopaedic and I’ve split the non medical term yuppie flu off into the controversies section.

All of which I think leaves the sub article redundant. If no one objects by next Monday I’ll put the redirect in place then – unless of course anyone else feels that it’s appropriate to go ahead before then.--In Vitro Infidelium (talk) 10:09, 11 January 2016 (UTC)


 * Redirect to Naming section of main article now in place.--In Vitro Infidelium (talk) 11:07, 18 January 2016 (UTC)

Pathophysiology – actions taken and proposal for redirect
In response to the suggestion made above to redirect some of the sub page articles Science Watcher noted the need to recover any relevant material to the CFS article before making the redirects, this was a simple task for the Alternate names article but more complex for the Pathophysiology article. Of the 154 references on the Path sub article the vast majority are unreplicated single study refs, or are now irrelevant – XMRV etc, or are woefully out of date; only around 20 or so strongly meet MEDRS – I’ve logged these at: https://en.wikipedia.org/w/index.php?title=User:In_Vitro_Infidelium/sandbox  along with comments of their value etc. If anyone feels that log belongs here, feel free to copy and paste.

The Pathophysiology sub article has a tag from 2009 ! identifying it as “too technical to understand”, having gone through the source list my conclusion is that the article is simply incomprehensible, unsalvagable and needs to be redirected with some urgency.

Proposal to redirect: If no one objects I will redirect the Pathophysiology article to the CFS page as of next Monday.

To ensure that material from the sub article is not lost, I’ve further updated both the Risk Factors and Pathophysiology sections of the CFS page, using those references from the sub article that were: --In Vitro Infidelium (talk) 10:15, 13 January 2016 (UTC)
 * relevant
 * not currently used
 * did not feature in the IOM bibliography
 * newer than a 1999 date.


 * Redirect to Pathophysiology section of main article now in place. --In Vitro Infidelium (talk) 11:08, 18 January 2016 (UTC)

Management section and Treatment sub article
I have made substantial changes to the Management section intro and to the CBT paragraphs – the GET and other paragraphs still require work. I had hoped to update the Treatment article in tandem with the main article however the scale of work has precluded that and I am now much less optimistic about the Treatment article being salvageable.


 * Fundamental problems:


 * On reading through the sources, while most adequately meet MEDRS very little of article text (main and sub) provided any recognition of the inconclusive nature of almost every resource – statements about limited scope, need for more research, uncertainty etc abound. The AHRQ(P2P) report acknowledges this global uncertainty with treatment research and it’s very difficult to square this with the main thrust of the Treatment article or the previous text of the main article.


 * Reference Trading – a great deal of the existing/previous text was contradictory, basically sources were placed in contrast to one another without any meaningful ordering.


 * Relevance of sources – I’ve used the CDC guide and NICE guidelines as the basis for the Management section intro – these give the real world interventions that most (English speaking) patients are likely to receive. The Management section shouldn’t be about contrary research assessments from fifteen years ago.


 * Side issues:


 * The IOM and AHRQ reports allow linking to sections via the refrence URL, I'm unclear what the MoS rules are for this but I've used link to sections rather than giving hard copy page numbers which often don't match or aren't given for the web reseource.


 * Link to the Treatment article - I've left this in place but my feeling is that while the Treeatment article remains as it is, the link ought to go and maybe a soft redirect put in place on the Treatment article page. --In Vitro Infidelium (talk) 14:42, 22 January 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 4 one external links on Chronic fatigue syndrome. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20100411035136/http://www.cbc.ca:80/health/story/2010/04/07/blood-donations-chronic-fatigue-virus.html to http://www.cbc.ca/health/story/2010/04/07/blood-donations-chronic-fatigue-virus.html
 * Added archive https://web.archive.org/20131014065501/http://www.donateblood.com.au/media-centre/latest-national-news/blood-service-updates-cfs-donor-policy to http://www.donateblood.com.au/media-centre/latest-national-news/blood-service-updates-cfs-donor-policy
 * Added archive https://web.archive.org/20100625094629/http://www.aabb.org:80/pressroom/Pages/cfsrecommendation.aspx to http://www.aabb.org/pressroom/Pages/cfsrecommendation.aspx
 * Added archive https://web.archive.org/20110106010232/http://www.mrc.ac.uk:80/Ourresearch/ResearchInitiatives/CFSME/index.htm to http://www.mrc.ac.uk/Ourresearch/ResearchInitiatives/CFSME/index.htm

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 04:22, 1 February 2016 (UTC)

General issues & improvements
Many parts of the article are unclear, contradictory, and/or confusing to laymen (who aren't used to navigating such hotly disputed ground in medicine). Examples:
 * "The fatigue of CFS is not due to ongoing exertion, is not much relieved by rest" (contradicted many times in article)
 * "...and is not due to any other medical condition" vs "It is unclear if these symptoms represent other associated conditions or if they are produced by CFS itself"
 * "Diagnosis is based on a patient's signs and symptoms" vs "testing is used to rule out other potential causes for symptoms. When symptoms are attributable to certain other conditions, the diagnosis of CFS is excluded" + "Current morbid obesity and active medical diseases need to be resolved and excluded before a diagnosis of chronic fatigue syndrome can be made" + "CFS is generally diagnosed by excluding other conditions"
 * "There is agreement that CFS has a negative effect on health, happiness and productivity but there is also controversy over many aspects of the disorder." (vague and unhelpful, sounds like HS/college essay introduction filler)
 * "Estimates of the number of people with the condition vary from 7 to 3,000 per 100,000 adults. About one million Americans and a quarter of a million people in the UK have CFS."
 * "Fatigue is a common symptom in many illnesses, but the fatigue experienced by persons with CFS is comparatively rare." vs "Certain medical conditions can cause chronic fatigue and must be ruled out before a diagnosis of CFS can be given." (and unclear by itself)
 * "These findings appear to be specific to ME/CFS and provide objective evidence" vs "There are no characteristic laboratory abnormalities"
 * ''"Important conditions and disorders to exclude are current/active major depression, schizophrenia, eating disorders such as anorexia nervosa and bulimia, bipolar disorder, alcohol abuse or other substance abuse.

I'd guess that editors may be too focused on POV/NPOV and end up adding content in parallel instead of thinking as authors/editors. If some statement seems unacceptably "biased", the correct response isn't to add a statement after that appears equally biased in the opposite direction. This isn't physics: they don't cancel each other out, they leave the reader confused and unsatisfied (or worse, convinced that one "side" has to be "right"). If there is no consensus among experts (or contradictory/insufficient evidence, etc) on some point, the article should just say that outright.

There's also a little too much defensiveness and victim's-advocacy-style statements (eg "this is a real disease, please don't joke about it", "these people suffer a lot, please don't make fun of them"), such as: This kind of stuff isn't particularly helpful or informative, should likely be stripped since it's dead weight for the article.
 * "Quality of life of persons with CFS can be extremely compromised."
 * "CFS affects a person's functional status and well-being more than major medical conditions such as multiple sclerosis, congestive heart failure, or type II diabetes mellitus"

Also, the article should probably note somewhere that CFS is generally (and notably) a wastebasket diagnosis in clinical settings: it's usually diagnosed not by specific criteria (signs/symptoms/tests), but by excluding all other possible diseases/disorders. The article even provides an example of this procedure: the CDC guidelines ("eliminate A,B,C,D,E,F,G,H,I,J, then eliminate every other disease known to modern medicine, then diagnose the patient with CFS").

Lastly, the article shouldn't mention specific early theories/treatments that are very far from widespread acceptance as if they're standard, especially not in an introduction that otherwise says "nothing is clear, everything is disputed, even the name":
 * "Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases. The medication rintatolimod may be useful for certain people." (as far as I know, CBT isn't specifically effective (ie it's only as effective for CFS as it is for any other disease, like placebo) and rintatolimod was denied approval for lack of efficacy)

-- bornLoser (talk) 11:16, 23 March 2016 (UTC)


 * Following on from your observations, can you now state what kind of specific changes to the article you recommend? It is not unusual for long articles to be internally contradictory in places. JFW &#124; T@lk  16:59, 27 March 2016 (UTC)


 * – welcome to the confused and confusing world of CFS. I have some sympathy for what you have said, however it is worth getting to grips with the history of the article before any new departure is embarked on. There has been considerable edit warring in the past and this is perhaps the first sustained period when progress has been made on building the article without there being significant displeasure on all sides.


 * In respect of the main points you make:

* Many parts of the article are unclear, contradictory, and/or confusing to laymen (who aren't used to navigating such hotly disputed ground in medicine).
 * CFS is not a simply defined condition, it has multiple iterations, is separately viewed from a number of different medical specialisms and is the subject of ongoing early stage research. There is only a limited amount of MEDRS and even those sources may be contradictory or cross purposed. Resolving these issues while mainting NPOV is a challenge, however your phrase hotly disputed ground overly emphasises the intensity of differences in medical opinion set in the broadest terms. There is a vociferous minority view with its locus in a small group of English psychiatrist and psyschologists together with a few researchers in Holland and Scandanavia, which takes an ambiguous position on the organic basis of CFS, however outside of the English NHS, a more expansive view of CFS appears to prevail.

''*I'd guess that editors may be too focused on POV/NPOV and end up adding content in parallel instead of thinking as authors/editors. If some statement seems unacceptably "biased", the correct response isn't to add a statement after that appears equally biased in the opposite direction. This isn't physics: they don't cancel each other out, they leave the reader confused and unsatisfied (or worse, convinced that one "side" has to be "right"). If there is no consensus among experts (or contradictory/insufficient evidence, etc) on some point, the article should just say that outright.''
 * Your analysis is at least in some respects correct – but consensus between editors has to operate, this is after all Wikipedia. Battlegrounds are often left in an untidy state, the question is how to to go about filling in the shell holes without starting another conflict. I haven’t gone through everyone of the quotes you take issue with but from memory all are taken from the best available MEDRS – and yes they are wanting, but are included either because they are representative of the current lack of knowledge or because they provide balance to other uncertain material. As regards consensus among ‘experts’ (who are they in a nascent field ?) I can only suggest you start reading the article ref list and see what other than “no one agrees about anything” could be said.

*There's also a little too much defensiveness and victim's-advocacy-style statements (eg "this is a real disease, please don't joke about it", "these people suffer a lot, please don't make fun of them"), such as: •	"Quality of life of persons with CFS can be extremely compromised." •	  "CFS affects a person's functional status and well-being more than major medical conditions such as multiple sclerosis, congestive heart failure, or type II diabetes mellitus"''' :This kind of stuff isn't particularly helpful or informative, should likely be stripped since it's dead weight for the article.


 * I think there is an issue of perspective here over what a Wikipedia article should be, my understanding is that WikiProject Medicine has developed a common format for articles and QoL would be part of the expected information. Likewise functioning is an important aspect of symptomology. The text used is a direct MEDRS quote.


 * Also, the article should probably note somewhere that CFS is generally (and notably) a wastebasket diagnosis in clinical settings: it's usually diagnosed not by specific criteria (signs/symptoms/tests), but by excluding all other possible diseases/disorders. The article even provides an example of this procedure: the CDC guidelines ("eliminate A,B,C,D,E,F,G,H,I,J, then eliminate every other disease known to modern medicine, then diagnose the patient with CFS").


 * Referencing the wastebasket diagnosis article doesn’t really help – on what basis is the CDC to be challenged on the grounds that it isn’t competent to define a diagnosis ? A diagnosis of exclusion is clearly less than satisfactory but that is what the NIH/CDC, NICE and the medical authorities in many other countries are agreed upon.

*Lastly, the article shouldn't mention specific early theories/treatments that are very far from widespread acceptance as if they're standard, especially not in an introduction that otherwise says "nothing is clear, everything is disputed, even the name": •''	"Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases. The medication rintatolimod may be useful for certain people." (as far as I know, CBT isn't specifically effective (ie it's only as effective for CFS as it is for any other disease, like placebo) and rintatolimod was denied approval for lack of efficacy)''


 * The quote is taken from the most recent and extensive MEDRS dealing with treatment. CBT forms the most extensive focus of treatment research for CFS, if the NIH’s P2P says it, it’s difficult to find a basis for not referring to it given the lack of MDRS opposing the P2P.


 * Despite having defended all that content, my own position is actually that a more parsimonious article would better serve the general reader. There is the issue though not just of the main article but of the sub articles that are still be addressed - see https://en.wikipedia.org/wiki/Talk:Chronic_fatigue_syndrome/Archive_19  CFS subarticles . The Pathophysiology and Alternative Names pages have been redirected to the main article but work on  Clinical descriptions, Treatment, and Controversies is still required with potential to add material to the main article before a redirect is put in place.


 * As JFW asks, perhaps you could makesome definitive proposals for content reduction, based on MEDRS and an attempt to retain balance. Before starting this I recommed you read https://www.ncbi.nlm.nih.gov/pubmed/25695122 and https://www.nice.org.uk/guidance/CG53  in full to get some grounding in the complexities of CFS. --In Vitro Infidelium (talk) 13:59, 2 April 2016 (UTC)

Content Parsimony
Discussion above has suggested some of the current article content and levels of what could be called overloading is not helpful to the lay person's understanding of CFS. As an attempt at approaching these problems I'm proposing the following as suitable for pruning:


 * Lead:


 * There is agreement that CFS has a negative effect on health, happiness and productivity but there is also controversy over many aspects of the disorder. Physicians, researchers and patient advocates promote different names[9] and diagnostic criteria, while evidence for proposed causes and treatments is often contradictory or of low quality.[10]


 * Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases. The medication rintatolimod may be useful for certain people.[8]


 * Symptoms:


 * Unlike the CDC's diagnostic criteria for CFS, the International Consensus Criteria for ME do not require the 6-month waiting period before diagnosis, noting that "No other disease criteria require that diagnoses be withheld until after the patient has suffered with the affliction for 6 months."[24]


 * Comorbidity


 * Many CFS patients will also have, or appear to have, other medical problems or related diagnoses. Fibromyalgia occurs in a large percentage of CFS patients between onset and the second year, and some researchers suggest fibromyalgia and CFS are related.[35] As previously mentioned, many CFS sufferers also experience symptoms of irritable bowel syndrome, temporomandibular joint pain, headache including migraines, and other forms of myalgia. CFS patients have significantly higher rates of current mood disorders than the general population.[36] Compared with the non-fatigued population, male CFS patients are more likely to experience chronic pelvic pain syndrome (CP/CPPS), and female CFS patients are also more likely to experience chronic pelvic pain.[37] CFS is significantly more common in women with endometriosis compared with women in the general USA population.[38]


 * Risk factors


 * A systematic review carried out in 2008 assessed primary studies which had examined demographic, medical, psychological, social and environmental factors as potential risks that might lead to the development of CFS. The review concluded that the lack of agreement between primary studies meant that none of these factors were helpful in identifying patients at risk of developing CFS/ME.[42]


 * Pathophysiology


 * A substantial body of evidence points to abnormalities in the hypothalamic-pituitary-adrenal axis (HPA axis) in CFS patients that may include mild hypocortisolism, an attenuated diurnal variation in cortisol, enhanced cortisol negative feedback, blunted HPA axis responsiveness and a hyperserotonergic state. It is unclear whether or not these disturbances play a primary role in the pathogenesis of CFS[50][52][53] or have a secondary role in exacerbating or perpetuating symptoms later in the course of the illness.[54]


 * Differential diagnoses


 * Fatigue and muscle pain occurs frequently in the initial phase of various hereditary muscle disorders and in several autoimmune, endocrine and metabolic syndromes; and are frequently labelled as CFS or fibromyalgia in the absence of obvious biochemical/metabolic abnormalities and neurological symptoms.[citation needed]
 * Multiple chemical sensitivity, Gulf War syndrome and post-polio syndrome have symptoms similar to those of CFS,[63][64] and the last is also theorized to have a common pathophysiology.[64]


 * Other


 * Evidence suggests that the medication rintatolimod may be useful in some people.[8] In 2012, however, the FDA considered the evidence for the safety or benefit of rintatolimod to be insufficient to approve its use.[78] Other treatments of CFS have been proposed but their effectiveness has not been confirmed.[79] Medications thought to have promise in alleviating symptoms include antidepressant and immunomodulatory agents.[80] The evidence for antidepressants is mixed,[81] and their use remains controversial.[82] Many CFS patients are sensitive to medications, particularly sedatives, and some patients report chemical and food sensitivities.[17] A 2005 meta-analysis concluded that CFS patients have a low placebo response, especially to psychological-psychiatric interventions.[83]


 * Prognosis


 * A 2014 systematic review reported that estimates of recovery from CFS ranged between 0 to 66% in intervention studies and 2.6 to 62% in naturalistic studies. There was a lack of consensus in the literature on how recovery should be defined. "Recovery" was often based on limited assessments, less than a full restoration of health, and self-reports with a general lack of more objective measures, which when used, did not find significant changes in physical activity. The authors suggested that patients were still avoiding post-exertion symptom exacerbation, and could be clinically improving to a limited extent or adapting to ongoing illness rather than recovering. It was recommended using stricter and more comprehensive definitions of recovery which capture fatigue, function, patient perceptions, and recovery time following physical and mental exertion.[86]


 * Social issues


 * A study found that CFS patients report a heavy psychosocial burden.[119] A survey by the Tymes Trust reported that children with CFS often state that they struggle for recognition of their needs or they feel bullied by medical and educational professionals.[120]


 * Social support


 * Individuals with CFS may receive a poorer quality of social support than in those with other illnesses. One study found that CFS patients reported an increased incidence of negative/unsatisfying interactions with family, friends, colleagues and doctors, when compared with healthy controls and breast cancer patients currently in remission.[121]

--In Vitro Infidelium (talk) 15:09, 3 April 2016 (UTC)

Deletions made 10th April. --In Vitro Infidelium (talk) 12:18, 10 April 2016 (UTC)


 * You seem to have deleted some important info from the article. What are your arguments for removing the HPA axis, prognosis and comorbidity sections, and the treatments from the lede? --sciencewatcher (talk) 16:36, 10 April 2016 (UTC)
 * Yes much of this was well supported. Therefore I am restoring some of it. Why remove an overview of treatment from the lead? Doc James  (talk · contribs · email) 19:14, 10 April 2016 (UTC)


 * Re: the CBT and rintatolimod sentence although the source is unquestionably valid the information provided is weak and touch weaselish (“can be beneficial in some cases”, “may be useful for certain people” ) and in the case of rintatolimod it is contradicted by the position of the FDA. If there was a source to quote which allowed some definition of which patients benefit from CBT, or and/or a source which allowed rationalisation of the IOM statement with the FDA position then the sentence would be more strongly encyclopaedic. But I don’t thinks its harmful where it is,  just that it’s a bit of the dead weight that Bornloser referred  to, anyway lets leave it in for now and await developments regarding CBT and Ampligen.


 * Bornloser made some valid criticisms one of which was that the article is overly technical and as such not of use to the general reader. The HPaxis stuff remains of hypothetical relevance no matter how many MEDRS present it as important. To make the HPAxis material encyclopaedic would require substantially more explication than is currently given, which in turn would imbalance the article further in terms of weight given to this one set of measures that may have no substantive relevance to the pathology of the disease/diseases being reported upon.


 * The question of what material belongs isn't just whether it is RS, it's whether it supports accessibility and understanding for the reader. What value is telling the reader that estimates of recovery are between 2.6% and 66% when there's no sound definition of recovery and such wide variation is itself meaningless or that CFS patients have a low placebo response when there is nothing meaningful to relate to that fact. It's been a long time since there was much focus on what the reader needs from the CFS article, and whole lot of concern about what editors want to present about the illness, Bornloser's term "deedweight" is a useful one and in some aspects at least, for the reader - less is more. --In Vitro Infidelium (talk) 15:00, 13 April 2016 (UTC)

Addition
This content "Studies do not however show an improvement based on objective measures ." is supported by an non pubmed indexed review as the journal has only existed for a few months.

This is primary sources ". A phase II double-blinded placebo-controlled study published in 2011 showed that 10 out of 15 patients experienced a moderate or major improvement from the drug Rituximab, compared to 2 out of 15 in the placebo group. A phase III follow-up study with 152 patients is ongoing in Norway. The study will be published in 2018." Doc James (talk · contribs · email) 14:28, 17 April 2016 (UTC)

CBT/GET Harms
Reporting of Harms Associated with Graded Exercise Therapy and Cognitive Behavioural Therapy in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome http://www.ncf-net.org/library/Reporting%20of%20Harms.pdf and A review on cognitive behavorial therapy (CBT) and graded exercise therapy (GET) in myalgic encephalomyelitis (ME)/chronic fatigue syndrome (CFS): http://www.ncbi.nlm.nih.gov/pubmed/19855350 were used as sources for two recent additions to the management section. The Bulletin of the IACFS/ME appears not to be peer reviewed, although personally I think Tom Kindlon's work is wholly valid the publication doesn't seem to meet the test of MEDRS - note the URL given in the recent edit was wrong - above is correct. If anyone has a sound argument for support of the Bulletin as MEDRS then I would support use of Kindlon in the Management section. Twisk (not Twist) and Maes is more problematic because although described as a 'review' it does not meet the the standard of MEDRS - quote from abstract: In this review we invalidate the (bio)psychosocial model for ME/CFS and demonstrate that the success claim for CBT/GET to treat ME/CFS is unjust. --In Vitro Infidelium (talk) 09:25, 27 April 2016 (UTC)

Problems of controversy
CFS and CFS controversy articles appear in the WP category medical controversies – there is no other somatic disease amongst those listed with the exception of Lyme Disease which is included because of the controversy related to the doubtful diagnosis of ‘chronic lyme’. There are no MDRS that support the categorisation of CFS as a ‘medical controversy’. The are a number of issues about CFS which have been historically controversial but to the extent that these have any relevance, they can/are/should be dealt with in the main article history section. As it stands the current controversies section is a wastebin category of largely irrelevant material. I propose to change the section to Media Controversies with the following text:

Media Controversy

''Media treatment of CFS has been controversial. In November 1990 the magazine Newsweek ran a cover story on CFS which although supportive of an organic cause of the illness also featured the term Yuppie Flu. Reflecting a stereotype that CFS mainly affected yuppies, the implication was that CFS was a form of burnout. Use of the term Yuppie flu is considered offensive both by patients and clinicians Books which cast CFS as being either a social construct or wholly psychiatric in origin have added to the media presentation of CFS as a controversial issue. The mistaken linking of CFS with the XMRV retrovirus was widely reported in the news media, with particular attention being given to the issue of blood donation. ''

Once this edit is completed I also propose to redirect the Controversies sub article to the new section. As previously discussed the Controversies sub article is beyond salvation and has been tagged since 2008 without any substantial improvement.--In Vitro Infidelium (talk) 12:43, 6 April 2016 (UTC)


 * i'm not sure why you think it's just the media treatment of CFS is controversial. The entirety of CFS is controversial. This should be even more crystal clear at the moment, with the current continuing controversy over the PACE trial. (That's not just the media). I support moving the controversy article into the main article, but I don't think it should be renamed "media controversy" --sciencewatcher (talk) 16:38, 10 April 2016 (UTC)


 * Controversy is a problematic construct, in WP terms it borders on WP:WEASEL and needs to be approached very cautiously. As a blanket term it is unencyclopaedic and if we are going to deal with controversy related to an illness we should be clear where the controversy is located and how it is supported by RS. In the case of a medical controversy I would argue that MEDRS should be the defining source, so there needs to be something in an MEDRS that defines what the controversy is – it is not enough to have some medic or researcher say x is controversial, there should be clear definition of where in medicine the controversy is located. For example, is a diagnosis controversial ? or a treatment ? or is there controversy over the basic science of an illness, as with (historically) the capacity of H.pylori to survive or not in human stomach acid ? The point about medical controversies is that they tend to resolve as research progresses, while political, media, social etc controversies do not, although they may fade from general interest. The case of PACE is instructive, the controversies over PACE are not controversies over CFS, they are about the problems related to a single study and a small group of researchers with a specific perspective. In science controversy isn’t about mere difference of perspective – different groups of researchers are in energetic argument with other groups all the time, for there to be medical/scientific controversy there needs to be more than the usual fighting of corners, there needs to be an exceptional divide in scientific perspective. I do not see where CFS is controversial in medicine, there is no denial of its existence as an illness by any major health body (at least as far English language sources), there are well recognised processes of diagnosis (Fukada, NICE etc), there is broad acknowledgment that treatment options are limited in both number and utility, and there is wide acceptance that causation is currently unknown. So where is the medical controversy ?


 * Media controversies aren’t necessarily easier to deal in terms of avoiding WP:WEASEL but the source of the controversy is usually clear – i.e it’s the publication the kicked off of featured an exceptional claim that is not supported by any broader authority. In the case of media controversies it’s enough to simply identify the publication and the relevant content and leave it to the reader to make further assessment. The Yuppie Flue issue and the reviews of the books by Showalter and O’Sullivan are media controversies with CFS at their core –negative naming, denial of existence as a real disease and exclusively locating the illness as exclusively psychiatric. --In Vitro Infidelium (talk) 15:04, 13 April 2016 (UTC)


 * If you look at the controversies section you'll see ample MEDRS sources for the medical controversy. The organic/psychiatric controversy is basically what it all boils down to. The fact that you yourself see psychiatric illness as "not a real disease" means that you yourself are part of the problem as well... Perhaps you meant to say "denial that it is an organic illness"? Most researchers and doctors also have varying theories about psychosomatic/psychiatric/organic causation of CFS, and many doctors and researchers themselves have dubious, bigoted and incorrect ideas about psychosomatic and psychiatric illness, which adds to the problem. --sciencewatcher (talk) 17:07, 13 April 2016 (UTC)
 * Some controversies do exist and we should describe them. Doc James  (talk · contribs · email) 19:13, 13 April 2016 (UTC)


 * @Science Watcher. I suspect I have far more direct experience of the reality psychiatric illness than you are ever likely to, so please don't presume to label me as some kind of psychiatric denier. I did not as you suggest, mean to say that "denial that it is an organic illness", Showalter was not making a case for psychiatric illness but rather for illness as a proxy for a state of non legitimate emotional failing. O'Sullivan's case appears to be that CFS and various other conditions that don't fit her neurology model are indeed "all in the mind". I have read very thoroughly the Controversies page and there is not a single MEDRS that supports defining CFS as a medical controversy. Wessely, White and perhaps one or two others of the UK BPS school, do describe CFS as controversial but without other specialisms and research perspectives also using the epithet it simply leaves controversial as a BPS characteristic related to a perspective that is peculiar to the BPS adherents. There is no explantion anywhere why they use it, and so it's not even definable in BPS terms. In any case WP:WEIGHT applies.


 * Showalter most certainly does not propose that CFS is "proxy for a state of non legitimate emotional failing". Did you actually read her book? Her view is that CFS is a psychological illness that has real symptoms and real effects, and that it is a "real" illness. She criticizes people like Cheney who believe that psychological illnesses aren't real. Read p130 and you'll see this very clearly spelled out by her. We're getting off topic though.
 * There seem to be quite a few MEDRS refs supporting the controversy. Hooje for example says in the abstract "There is much controversy as to whether chronic fatigue syndrome is a physical or a psychological illness". To be honest I'm a little baffled as to how anyone could claim that CFS isn't medically controversial unless they were living under a rock. --sciencewatcher (talk) 19:23, 15 April 2016 (UTC)


 * @Doc James. What controversies ? Having a small group of one medical specialism in one country, amongst whom there is a particular perspective that is itself intimately connected to the definitional status of a subsection(Liaison Psychiatry) of that specialism within the service structure of the national health provider, isn't a case of medical controversy. It's just a parochial (by sub specialism, by health provider, by country (England not even whole of the UK)variance in medical opinion. Of course if there are MEDRS supported controversies WP should cover them but at present the material simply isn't there. Quite the opposite, there is fundamental agreement on diagnosis, on the limitations of available treatment and on the lack of identification of causal or pathological processes. What else is there about which an illness could itself be controversial ? --In Vitro Infidelium (talk) 14:13, 15 April 2016 (UTC)


 * The Hooje reference is a quarter of a century old. For encyclopaedic purposes we need to distinguish what is historic and (in balanced form) current. Perceptions in medicine change, without recent, appropriate and balanced MEDRS which explain how CFS is controversial (diagnosis, pathology, management etc), there is no basis to describe it as such, to do so is OR etc. In Vitro Infidelium (talk) 11:21, 19 April 2016 (UTC)
 * Perhaps it's time to look for some new references then. If you do a >=2012 search on gscholar there are lots of MEDRS references for CFS controversial. --sciencewatcher (talk) 15:14, 19 April 2016 (UTC)

Clear all the false positives and we are left with non MEDRS or just unsupported statement/opinion from psychiatrists/psychologist dominated publications. One can't have controversy where just one section of the disputing population claims controversy exists - and provides no evidence for where the controversy lies - is it diagnosis, is it on the limitation of management/treatment options, is it on the absence of identified pathology ? --In Vitro Infidelium (talk) 09:38, 27 April 2016 (UTC)

CFS a mitochondrial disorder
There is good evidence now that, at least in some individuals, CFS involves dysfunction at the mitochondrial level in the recycling of ATP-ADP. Please add mention of that fact to the discussion of etiology here. (I am not an expert, or I would undertake the task myself.)

Thanks KC 00:19, 28 May 2016 (UTC)

KC 00:19, 28 May 2016 (UTC) — Preceding unsigned comment added by Boydstra (talk • contribs)


 * I haven't seen any research myself, except a study from Myhill which hasn't been validated. (I think another team is currently validating their findings).
 * I'm not sure how mitochondrial dysfunction could really make sense for CFS, as it would presumably result in permanent fatigue rather than post-exertional fatigue. --sciencewatcher (talk) 00:50, 28 May 2016 (UTC)

UK Situation in General
The material on the 'British situation' has undue weight in terms of it's presence on this page that does nothing to improve understanding of the illness, while the symptoms section is little more than a list and very hard to read. This should be moved into the Controversies page. Leaving it here creates a fork lacking balance and lacks a Neutral point of view WP:NPOV.

It also needs to be much more 'Journalistically sound' by properly drawing together the full impact of those targeted and the limited extent of that activity. — Preceding unsigned comment added by Leopardtail (talk • contribs) 12:57, 11 May 2014 (UTC)

Associated with gut bacteria differences
Could mention "Chronic fatigue syndrome is in your gut, not your head" (Cornell University, June 27, 2016, Science News, sciencedaily.com) Says "Now, for the first time, researchers report they have identified biological markers of the disease in gut bacteria and inflammatory microbial agents in the blood." - Rod57 (talk) 12:15, 7 July 2016 (UTC)

Keep the pathophysiology section on topic please
The pathophysiology section is not the place to speculate about causes or patient beliefs. It's the place where you describe the things that have been documented to be wrong in patients bodies. I have made an edit to remove the inappropriate paragraph by sciencewatcher. — Preceding unsigned comment added by Abmayer (talk • contribs) 19:07, 7 July 2016 (UTC)

As I'm writing this, sciencewatcher has reverted my edit, with the justification "just because you disagree with it doesn't make it "off-topic". This dispute is easily solved by consulting a dictionary:

The Medical Dictionary for the Health Professions and Nursing defines pathophysiology as '''1. The study of structural and functional changes in tissue and organs that lead to disease. 2. Derangement of function seen in disease; alteration in function as distinguished from structural defects.''' http://medical-dictionary.thefreedictionary.com/pathophysiology

Clearly, it's not "just me". A paragraph that deals with patient beliefs about their illness is severely off topic in the pathophysiology section. I'm going to remove the paragraph again. Please accept that this is not the right place for this information. Abmayer (talk) 20:31, 7 July 2016 (UTC)


 * It's in the pathophysiology sections of both the Buchwald and Wyller reviews we're using in the article, so that does seem to be the correct section. However if you feel it should go into a different section, please suggest another section rather than just deleting it. --sciencewatcher (talk) 21:03, 7 July 2016 (UTC)


 * I have checked both the Buchwald and Wyller reviews and neither of them discusses patient beliefs in their pathophysiology sections. I think you are POV pushing. This seems to be based on the misinterpretation of the correlation between physical attribution and worse outcomes. It's just a correlation that tells you nothing about causality. I think speculation of this kind adds little value to an article, so I won't support inclusion of it.Abmayer (talk) 05:42, 8 July 2016 (UTC)


 * I don't think this is controversial content. This is very much about pathophysiology and as long as the sources support the notion that beliefs may influence symptoms it is perfectly alright. Can I suggest that you stop removing the paragraph? The most effective way of challenging content is recommending changes (rather than deletion) and forming consensus around these. Thanks. JFW &#124; T@lk  09:32, 8 July 2016 (UTC)


 * Addendum: if anything, the title of the section should be changed to "Mechanism" rather than "Pathophysiology" (see this MOS entry) to avoid further disputes. JFW &#124; T@lk  09:33, 8 July 2016 (UTC)


 * I have also read the Buchwald review and it reports only correlational findings, ie. belief in physical cause is associated with greater number of symptoms and impairment, but as everyone knows, correlation doesn't imply causation. That could just as well mean that illness causes belief to be ill. I agree that a group of psychiatrists believes that CFS is a mental disorder of false illness beliefs, but this should not be presented in a manner that overstates the evidence or under a pathophysiology, or even mechanism, section, because it's unproven speculation. Should we really start making a list of all the proposed causes of CFS? It would be very long! Abmayer (talk) 11:33, 8 July 2016 (UTC)


 * I suspect a list of "causes" would be impossible because causation has not really been proven for any of those.
 * To be honest, I am pretty sure that illness perception contributes to symptoms and symptom interpretation in all forms of illness, including the classical examples (heart attacks etc). CFS is not necessarily exempt from that phenomenon.
 * I have rephrased the paragraph to state that this is a matter of association rather than implying causation. Causation will probably never be proved: one would need to design a prospective trial with widespread adjustment for confounders with a high risk of residual confounding. JFW &#124; T@lk  12:54, 8 July 2016 (UTC)


 * I continue to hold the view that this paragraph is completely inappropriate for a pathophysiology or mechanism section, until there is actually evidence that beliefs truly cause symptoms. It would be just as inappropriate to list the various pathogens associated with CFS here, based on the speculation that they are causing symptoms. Just like with the false illness beliefs hypothesis, one can easily find researchers proposing these to be cause of symptoms, but until concrete evidence emerges these are out of place in a pathophysiology section. There seem to be double standards being applied here: high standards for biomedical research, low standards for psychological research. I hope that editors can weigh in on this issue. Abmayer (talk) 13:54, 8 July 2016 (UTC)


 * The fact is that "illness beliefs" were put forward as the basis for CBT in the large PACE trial, so clearly there is weight behind these views. None of the things in the pathophysiology have "concrete" evidence. All of them are somewhat preliminary. The only real criteria is that reviews mention the factors as possible/probable factors. If there are different standards then we should mention that, but what is the evidence? For example, most of the immune studies and HPA axis are somewhat disputed - not all studies show these abnormalities in patients. Whereas for the psychological factors, are there any studies that have looked into them and not found them to be factors? I get the feeling that you just don't like them. That's fine, and many CFS patients feel the same way, but not a reason for excluding them. I don't think illness beliefs are central to the illness myself, but the whole point of wikipedia is that it represents the state of the science and significant viewpoints of researchers, not our opinions. --sciencewatcher (talk) 14:26, 8 July 2016 (UTC)


 * I think the problem is that the added paragraph is not complete and leads to misinterpretations. It should summarize the conclusions of the references and it does not. I will respect the paragraph, but I will add a couple of sentences, supported by the source already present (see full text here), to complete and give neutrality. Best regards. --BallenaBlanca (talk) 19:51, 9 July 2016 (UTC)
 * Thanks for your edit. I've just tweaked it a little bit for correctness. Both patients and relatives tend to give somatic (physical) rather than psychological explanations. I think you and the other editor might have been confused by the word "somatic". It means "physical" and not "psychological" (soma = body). If you don't believe me, have a look at "Causal attributions for somatic sensations in patients with chronic fatigue syndrome and their partners" where this is explained clearly. (There's also a hint in Afari where she says "relatives also tend to attribute") --sciencewatcher (talk) 21:11, 9 July 2016 (UTC)


 * The paragraph is still off topic for the pathophysiology and now it is also far too long. It should summarize information and be moved into an appropriate section. Abmayer (talk) 06:26, 10 July 2016 (UTC)


 * The paragraph isn't long at all. It tackles a really thorny issue, namely as to whether patients' beliefs contribute to the prognosis. As I explained above that is definitely not "off-topic" and it seems to be supported with reasonable evidence. It is already in summary style and doesn't need to be moved anywhere.
 * Of course the paragraph would be unacceptable to those who believe that ME/CFS is always caused by persistent physical illness. You might hold these views, and it might help if you declared that upfront because the discussion would be helped by this. JFW &#124; T@lk  12:40, 10 July 2016 (UTC)


 * I wouldn't object if the article contained a paragraph describing how psychiatry views CFS. This is a significant viewpoint. Rules say it should be included. There is however no evidence that belief to be ill causes the physical symptoms of CFS. This is merely how some psychiatrists have decided to interpret the correlation between these two factors. Wikipedia has clear rules on this: opinions must not be presented as facts. https://en.wikipedia.org/wiki/Wikipedia:Neutral_point_of_view/FAQ#Assert_facts.2C_not_opinions Abmayer (talk) 16:13, 10 July 2016 (UTC)
 * I think the current wording makes sense in the pathophysiology section and is an accurate reflection of the sources. —PermStrump  ( talk )  17:27, 10 July 2016 (UTC)


 * The paragraph that talks about somatic perceptions doesn't present it as a fact, but says "Although it has been suggested that somatic attributions may be a risk factor for the development of chronic fatigue syndrome (157), at the very least, they probably exacerbate the illness and lead to greater disability." So I will adjust to the source. Also, IMO this sentence "Some individuals with CFS firmly reject any psychological involvement and believe strongly that their condition has a physical cause." is redundant and doesn't reflect the conclusions/objetive of the source that support it, which is aimed at analyzing the placebo effect of psychological interventions. Best regards. --BallenaBlanca (talk) 19:22, 10 July 2016 (UTC)
 * That sentence from the placebo study was already in the pathoaphysiology article, so I just copied it in here. I think it was added as an explanation...it's an important point that CFS patients don't like a psychosomatic view of the illness. Whether or not it should be placed right there is another point. Perhaps it would be better in the controversies section, and maybe use a review for the source. --sciencewatcher (talk) 21:25, 10 July 2016 (UTC)


 * The fact that Medicine can not find an explanation does not mean that the disorder is psychosomatic. It is logical and intelligent attempt to find a cause because it means that may be found a treatment, but the suffering of a patient whose condition is not identified, for which there is not an effective treatment, prognosis is unknown, and health professionals do not know how to handle or even do not listen to the patient, is greater, of course. I have added a sentence from NICE . Best regards. --BallenaBlanca (talk) 23:14, 10 July 2016 (UTC)


 * The whole concept of psychosomatics is a little bit outdated. It is a reality that you cannot separate illness, whether "psychiatric"/psychological or somatic, from the way it is perceived by the person experiencing it. That is not a view limited to the field of psychiatry, and discussing these issues in the "pathophysiology" section is not a matter of "how psychiatry views CFS" but "how CFS works" full stop.
 * There are numerous conditions that have no satisfactory pathological explanation, or where the underlying science is the matter of controversy. That should not deter clinicians for focusing on the strategies that have been proven to work, and identifying issues that might interfere with effective recovery. JFW &#124; T@lk  14:24, 11 July 2016 (UTC)

Robert Naviaux et al "Metabolic features of chronic fatigue syndrome"
This study appears to be significant and has found a pattern of metabolic abnormalities in CFS patients http://www.pnas.org/content/early/2016/08/24/1607571113.full  The Open Medicine Foundation (that funded the work) announced in July a replication study was underway http://www.openmedicinefoundation.org/2016/07/06/metabolomic-and-chronic-fatigue-syndrome-robert-naviaux-md/  Most media coverage has been of very poor quality, but this article is perhaps the best http://www.economist.com/news/science-and-technology/21706241-new-test-may-diagnose-mysterious-illness-and-also-help-explain-it-blood  The Open Medicine Foundation is a very small CFS research initiative but punching well above its weight - its scientific advisory board has three Nobel laureates and six National Academy of Science members http://www.openmedicinefoundation.org/scientific-advisory-board/  It is led by Stanford's Ronald W. Davis (whose son is very sick with CFS) https://en.wikipedia.org/wiki/Ronald_W._Davis

I do not know the Wikipedia rules (so have not edited the main page) and am not sure if study replication is required or any other bars met for study quality or size, but even if this is not appropriate to be included now, it seems useful to post in this talk page as probably in time this topic of metabolic defects will become something the page should cover, especially given the common symptom of post exertional malaise where patients get much sicker after minimal exertion, and what Mark VanNess has found to be a totally broken aerobic metabolism than cannot be trained. Where we have good sources, this sort of information is crucial because it gets right to the core of why patients are so infuriated by exercise therapies (patients surveys typically show high levels of dissatisfaction and claims of harm from these treatments). Kicsinyul (talk) 07:59, 5 September 2016 (UTC)
 * Please read WP:MEDRS. We should be using secondary sources. Best Doc James  (talk · contribs · email) 09:57, 5 September 2016 (UTC)

AHRQ addendum
The addendum is talking about CDC vs Oxford criteria, so the addition to the article isn't quite correct. Can the OP please fix? --sciencewatcher (talk) 03:15, 20 August 2016 (UTC)

Also, it should go into the treatment article, and you should give a proper summary of their conclusions on CBT and GET. --sciencewatcher (talk) 04:08, 20 August 2016 (UTC)

Thank you sciencewatcher

I am currently unable to make edits so please feel free to make whatever amendments / deletions you require to my contribution on this subject

C7762 (talk) 18:07, 20 August 2016 (UTC)


 * I don't have time either right now. The main problem is that your current edit completely misrepresents the source. So I think it's either a case of you (or someone else) rewrite it, or it will probably get reverted. --sciencewatcher (talk) 23:46, 20 August 2016 (UTC)

The AHRQ noted that the evidence base on CBT and GET depends heavily on studies that use the Oxford case definition of CFS, which only requires the symptom prolonged fatigue to qualify for a diagnosis of CFS. The AHRQ describes the Oxford case definition as the least specific case definition, with high risk of including patients that have a different illness, or an illness that resolves spontaneously over time. The purpose of the addendum was to assess the impact of the Oxford case definition on the conclusions, and to assess the impact of separating studies of CBT from other counseling and behavioural interventions. Table 5 and 7 show the change in strength of evidence. When Oxford case definition studies are excluded, there is insufficient evidence that GET results in improvement on any outcome (these being function, fatigue, employment, quality of life, global improvement). With CBT, low strength of evidence for improved function is downgraded to insufficient evidence, and low strength of evidence for global improvement is downgraded to insufficient evidence, strength of evidence regarding employment remains insufficient, strength of evidence on fatigue remains low, strength of evidence that CBT does not improve quality of life remains low. Abmayer (talk) 17:50, 23 August 2016 (UTC)

As far as I know, the Oxford case definition is not actually used by any health agency or in clinical practice. It's a case definition used by some researchers that view CFS as being equivalent to symptom prolonged fatigue, but that's not how the rest of the world has defined CFS. I don't see any misrepresentation of the source material here. On the contrary, this addendum shows that if anything the misrepresentation lies in citing Oxford case definition based material as if it broadly applied to CFS patients when said patients are typically diagnosed with the CDC case definition. Abmayer (talk) 17:49, 23 August 2016 (UTC)


 * That's my point. The edit didn't mention any of that, or even mention Oxford criteria. As such, it severely misrepresented the Addendum. --sciencewatcher (talk) 17:52, 23 August 2016 (UTC)


 * It doesn't need to mention that because the CDC definition is the widely used, default definition as one could put it, while the Oxford definition is a fringe definition that multiple US health agencies have described as flawed and that isn't actually used in clinical practice. Conclusion based on studies using the CDC criteria can be generalized to the broader patient population because said population will consist mainly of patients diagnosed according to the CDC definition, but the same cannot be said about the Oxford definition. The statements in this article describing CBT and GET as helpful based on the evidence need to clarify that this applies only if a fringe definition is used. Abmayer (talk) 18:21, 23 August 2016 (UTC)

Hi - I have removed my edit until I or someone else can put together a bit more which covers it in more detail and hopefully a consensus can be reached. The whole reason for the AHRQ addendum is that the validity Oxford criteria is being questioned by AHRQ and several US government agencies which is the background context this edit needs.

Thank you C7762 (talk) 18:35, 23 August 2016 (UTC)

Link to the report. The addendum is in the beginning. Abmayer (talk) 15:15, 24 August 2016 (UTC)

Thanks Abmayer , I add reference to executive summary of a symposim? by the National Institutes of Health in 2015 that concluded that "continuing to use the Oxford definition may impair progress and cause harm. Thus, for needed progress to occur we recommend (1) that the Oxford definition be retired, (2) that the ME/CFS community agree on a single case definition (even if it is not perfect), and (3) that patients, clinicians, and researchers agree on a definition for meaningful recovery. "

The AHRQ then performed the re-analysis and addendum.

C7762 (talk) 18:30, 24 August 2016 (UTC) References:


 * Comment: Let us not forget that the Oxford was a Simon Wessely magination. So why waist time in even discussing it, other than for historical purposes.--Aspro (talk) 13:04, 30 August 2016 (UTC)

Thank you Aspro for your input. From my point of view it is important to record the 'historical' facts on this matter using references to sources that carry weight and give balance to all viewpoints. The "Wessely" hypothesis/theories are currently being applied to research in fatigue in :- Cancer, end stage kidney failure, multiple sclerosis and stoke patients so it is also relevant going forward. - "What's past is prologue" - I look forward to your further input. Thanks C7762 (talk) 08:56, 10 September 2016 (UTC)

Pace trial
The 5 million pound PACE Trial tested the model that the main maintaining factors for symptoms and disability in CFS are the patients beliefs and physical de-conditioning.


 * From Trial CBT Manual: Just as there are many factors involved in triggering CFS/ME, there are also many factors that are involved in sustaining it. According to this model, the symptoms and disability of CFS/ME are perpetuated predominantly by unhelpful illness beliefs (fears) and coping behaviours (avoidance). These beliefs and behaviours interact with the participant’s emotional and physiological state and interpersonal situation to form selfperpetuating vicious circles of fatigue and disability. Although it is acknowledged that lack of physical fitness may play a part in exercise induced symptom production, physical fitness is not central to this conceptualisation of the syndrome.


 * From Trial GET Manual: GET assumes that CFS/ME is perpetuated by deconditioning (lack of fitness), reduced physical strength and altered perception of effort consequent upon reduced physical activity.

On the 16th September 2016 Queen Mary University London released the analysis of improvement for the trial according to the original protocol (they have not released the Recovery figures according to original protocol) showing that CBT & GET did not result in the Lancet article improvement rate of ~60% but ~20%. The improvement rate for doing nothing was 10%. The 10 point point difference could easily be caused by the reported weaknesses and biases in the study. From this we can read that the trial failed to prove that CBT and/or GET are helpful in improving CFS and therefore failed to prove the underlying model. I am looking forward to hearing others views on this but my feeling is that this section will need to be altered drastically and possibly the psychosocial model for CFS put in an historical context. On the 17th September 2016 QMUL released anonymised trial data under a FOI request as ordered to by a first-tier tribunal. Independent analysis is expect shortly. Thanks. C7762 (talk) 11:39, 12 September 2016 (UTC)

Please find references for above

PACE Trial CBT treatment manual (page 18)

PACE Trial GET treatment manual (page 20)

Links to Statement of release of data and to Re-analysis according to original protocol available on this page http://www.wolfson.qmul.ac.uk/current-projects/pace-trial/#news

Additional reading:-

QMUL's defence of criticisms of PACE trial

Critique of PACE trial

thanks C7762 (talk) 08:16, 13 September 2016 (UTC)

Analysis (of PACE trial data according to original protocol) of Recovery rates for Graded Exercise Therapy and CBT reveals no statistical significance over specialist medical care. The Lancet article reported recovery rates of 22% for both GET and CBT - this was based on a mid-trial revised protocol where "13% of participants at baseline simultaneously met the trial eligibility criteria for ‘significant disability’ and the revised recovery criteria for normal self-reported physical function". Where as analysis according to the original protocol reveals recovery rates of 7% (CBT) & 4%(GET) with no statistical significance over 3%(SMC). . Thanks C7762 (talk) 09:59, 24 September 2016 (UTC)

In response the PACE trial Authors argue against open trial data. C7762 (talk) 10:45, 24 September 2016 (UTC)

PACE trial tribunal ruling (QMUL v the IC and Matthees)
The UK First Tier Tribunal has decided that Queen Mary University London should make PACE trial data available to Mr Matthees, who requested it under the freedom of information act. The tribunal decision contains some interesting information that might be relevant to the CFS page or some sub page.

During the hearing, Prof. Chalder admitted that no threats have been made against researchers. Back in 2011 newspapers reported that researchers were facing death threats by militants for conducting psychological research into CFS.

She also admitted that a Cochrane review of exercise therapy for CFS was not independent since PACE trial authors were on the review panel.

The tribunal further concluded that there was no evidence of an organized campaign of harassment against researchers or PACE trial participants.

Abmayer (talk) 16:45, 16 August 2016 (UTC)


 * Is this relevant for the article here? There is evidence of harassment of other CFS researchers. As for the Cochrane review, Cochrane has strict protocols to ensure that people don't promote their own research when doing reviews. Or were you referring to a different standard of independence? JFW &#124; T@lk  22:54, 16 August 2016 (UTC)

It is significant enough for the BMJ to have an article about the Judgement and the online version of "The i" newspaper (formerly part of the Independent group) has an article on the tribunal judgement that supports the proposed edit by Abmayer (talk).This can be considered a reliable secondary source, its editor is Oliver Duff and the reporter is Paul Gallagher.

It is also relevant to this section as

1. The eventual release of the data as ordered by the court will allow independent scrutiny of the trial to accept or reject the findings and conclusions of the Pace Researchers. This will once and for all put an end to the Pace trial controversy. I know the Cochrane review is often cited as independent but one of the senior Pace researchers stated in her evidence to the court that in her opinion it was not. Whether it is or not - there is still the perception that it is not independent and this adds to the controversy.

2. the Pace researchers had the opportunity to provide evidence to the court about the harassment (it was part of their defence for not releasing the data) and the court found there was no evidence except for a bit heckling. JFW you argue there is evidence of harassment - please provide references to evidence that the Pace legal team and witnesses did not have access to when presenting the case to the court.

I suggest the following 1st draft :-

Judgement in a First-tier Tribunal (20-22 April 2016) where PACE trial researchers had the chance to present evidence use expert witnesses and be cross examined stated "assessment of activist behaviour was,in our view, grossly exaggerated and the only actual evidence was that an individual at a seminar had heckled Professor Chalder". Judgement in this case ordered the release of the PACE trial anonymised data under the Freedom of Information Act noting "there could be legitimate concerns that they wish to suppress criticism and proper scrutiny of their trial". Approximately £200,000 was spent on legal fees for this appeal by the appellant to try to prevent the release of the anonymised data. . The appellant Queen Mary University London issued a press statement stating "This has been a complex case and the Tribunal’s decision is lengthy. We are studying the decision carefully and considering our response, taking into account the interests of trial participants and the research community."

a bit need to be added about the main argument for not releasing the data that it was not anonymised enough which the court threw out. Thanks for considering my input.

C7762 (talk) 18:22, 23 August 2016 (UTC)

Hi,

Below is my final draft -I have tried to be concise and balanced in the post. I hope that there will be feedback on its content and also layout and a concensus will be reached. When consensus is reached or if I have had no input by the 20 September 2016 (implied consensus) I will post. It is relevant in several pages that talk about the Pace trial and CFS controversies I would like some feed back on where I should post it and where links to it should be posted - otherwise I presume it should be put on 2 or 3 different pages. I hope for your feedback so I we can get the edit right.

Judgement in a First-tier Tribunal (20-22 April 2016) ordered the release of the PACE trial anonymised data under the Freedom of Information Act noting "there could be legitimate concerns that they wish to suppress criticism and proper scrutiny of their trial". .The Tribunal rejected the appellant Queen Mary University London claims that releasing data would breach patient confidentiality, might jeopardise follow up studies and damage the reputation of the college. . The tribunal found by a majority that the data “has been anonymised to the extent that the risk of identification is remote.”. PACE trial Researchers had the chance to present evidence use expert witnesses and be cross examined also argued that their was extremist activists in the CFS community, the Tribunal judged "assessment of activist behaviour was,in our view, grossly exaggerated and the only actual evidence was that an individual at a seminar had heckled Professor Chalder". Queen Mary University London issued a press statement stating "This has been a complex case and the Tribunal’s decision is lengthy. We are studying the decision carefully and considering our response, taking into account the interests of trial participants and the research community.". Alem Matthees the second respondent in the case is quoted as saying "This case ended up costing me greatly in time, energy, and health (currently bedridden) … but false or misleading claims of recovery or remission from debilitating illness simply have no place in the scientific literature.”

The following may or may not be worth adding (comments please):-

Approximately £200,000 was spent on legal fees for this appeal by the appellant to try to prevent the release of the anonymised data. .

Jo Edwards,Professor emeritus, UCL said "If scientific interpretation is poor it deserves no protection. If it is good it needs none."

Although part of judgement was a majority of 2 to 1. The verdict as a whole was unanimous.

QMUL argued that they did provide data to other researchers "The PACE trial was carried out according to the regulatory framework for UK clinical trials, which aims to ensure that trial participants can be confident that their information is only ever used according to their consent, and that their data is only shared under obligations of strict confidentiality." "QMUL’s appeal against the Information Commissioner argued in favour of controlled and confidential access to patient data from the PACE trial. QMUL has shared data from the PACE trial with other researchers only when there is a confidentiality agreement in place and an agreed pre-specified statistical plan for data analysis."

But in evidence they also argued that the researchers they provided the data to were not independent. and the court ruled that if they had provided the data to other researcher they would have been in breach of data protect laws.

Thanks - and looking forward to constructive feedback.

C7762 (talk) 00:03, 1 September 2016 (UTC)

The text looks fine to me. I think it is valuable to link to the February letter from (I think 43) scientists to the Lancet asking for an independent review of the PACE trial data, including academics from Stanford, UCL, DePaul, Berkeley and Columbia. This letter was referred to in the tribunal decision which stated "The identity of those questioning the research, who had signed an open letter or supported it, was impressive". http://www.virology.ws/2016/02/10/open-letter-lancet-again/ (interestingly since sending it to the Lancet, editor Richard Horton offered to publish a letter from the scientists, they wrote one and submitted it to the Lancet, the editor then rejected it - I suspect eventually there may be a page needed for the PACE study itself as events around this trial go beyond CFS research). Note Jo Edwards full name is Jonathan Edwards. I think the cost of the tribunal defence should be included as we know it is true (they confirmed the cost) and it speaks directly to the strong motivation of the university to not allow release of the trials data. The tribunal ruling is mentioned in The Times here (paywalled) http://www.thetimes.co.uk/article/chronic-fatigue-syndrome-could-be-the-body-trying-to-hibernate-3fzp2zgxv with a full-text version visible here http://www.meassociation.org.uk/2016/08/chronic-fatigue-syndrome-could-be-the-body-trying-to-hibernate-the-times-30-august-2016/ Kicsinyul (talk) 06:44, 3 September 2016 (UTC)

Thanks Kicsinyul for your comments.

I agree the letter should be incorporated within the context of "The identity of those questioning the research, who had signed an open letter or supported it, was impressive" ... "that strong public interest in releasing the data given the continued academic interests" which was given significant weight in the judgement. Strong Public and Academic interest is also reported in the secondary sources.

I also agree that cost of over £200,000 is significant and signifies 'the strong motivation' of withholding the data. Again this is reported in the secondary sources.

An interesting juxtaposition to the £200,000 is that previously QMUL turned down a FOI (not for the actual data) but for a re-analysis of the data according to the original specification on the grounds that it would cost more than £450. My current thinking that this £450 is not part of the core information - any thoughts any one?

Kicsinyul I will update the proposed entry with your suggestions and put it here before I post it.

Is there any other primary or secondary sources for any additional response from QMUL, the PACE trial or the Lancet that I have missed and is felt would be appropriate to incorporate. The judgement and secondary reporting is pretty damning, it is hard to find anything to include in the post in QMUL's favour?

C7762 (talk) 17:55, 3 September 2016 (UTC)

Kicsinyul I'm not aware of any response from the Lancet, or anything from QMUL apart from their press release that you refer to - the investigators and their allies have been oddly silent. However I am told this official PACE FAQ page has been updated recently, and it does provide responses to some of the criticisms of the trial http://www.wolfson.qmul.ac.uk/current-projects/pace-trial#faq  (the fact it has been updated recently is not evident from the page, but it is a primary source for responses to at least some trial criticisms). The Lancet editor Richard Horton asked the open letter signatories (asking for independent re-evaluation of PACE) to write a short letter to be published in the Lancet, alongside a response from the PACE trial authors. The open letter signatories did that, and the Lancet then (bizarrely) rejected it without explanation http://www.virology.ws/2016/08/29/once-again-lancet-stumbles-on-pace/ Kicsinyul (talk) 07:37, 5 September 2016 (UTC)


 * News:  Pace Trial data is being released as per Ruling. This proposed post will prob. turn into something much bigger once independent analysis of data. Already QMUL have admitted that original statistical plan is a third on reported improvement outcomes (~20% vs ~60%, with no treatment giving a 10% improvment) - have not released recovery rates as per original plan but the independent analysis will reveal all. C7762 (talk) 17:38, 9 September 2016 (UTC)

Please See Talk:Chronic_fatigue_syndrome for latest. Thanks C7762 (talk) 10:01, 24 September 2016 (UTC)

Copy of recent comment archived by bot
Talk:Chronic_fatigue_syndrome/Archive_19 was only added in September 2016 and it seems a shame to archive it (and hide it) so soon before others can see and comment on it. Would anyone object to the comment being copied into this live talk page so we can continue to comment on it ? - Rod57 (talk) 13:50, 28 September 2016 (UTC)


 * Hi @Rod57, I was also surprised that this was archived before a consensus could be achieved - I agree that it should be copied back to live - as well as the original peer reviewed article there is a substantial number of secondary references that the considers the research as a potential major breakthrough . On a wider note on such a contentious subject which requires discussion and consensus before changes can be made auto archiving after 3 weeks seems to be a bit quick. Thank C7762 (talk) 14:16, 28 September 2016 (UTC)


 * In Talk:Chronic_fatigue_syndrome/Archive_19, it was proposed that content be generated from a recently published primary source. That note was responded to, and there is never going to be a different response. That matter was done. Jytdog (talk) 14:51, 28 September 2016 (UTC)

@Jytdog there is quite a lot of secondary sources related to this, It was my intention to provide them for scrutiny but every thing takes time and auto archiving in 3 weeks does not provide a lot of time. Thanks C7762 (talk) —Preceding undated comment added 15:25, 28 September 2016 (UTC)
 * Please restrict your postings here to concrete content proposals and discussion about them. If you want to make a new proposal about the matters in PNAS report referenced in that section, that are sourced to MEDRS sources, please do so.  You can do that in a new section; there is nothing to act on in the archived section.  Thanks. Jytdog (talk) 15:50, 28 September 2016 (UTC)

"Onset" section on the Chronic Fatigue syndrome: Some thoughts
The main review article that’s cited in this section is reliable (citation [18]), but the current phrasing is lifted word-for-word from that review, so needs to be either in quotes or rephrased. Also, there’s a final statement at the end, which is taken word-for-word from an earlier, poorer quality source (a conference abstract). I suggest omitting that.

I also suggest adding a line from the main review article already cited about the findings of prospective studies. This approach compensates partially for the problem of recall bias in retrospective studies.

So here’s a suggested rephrase:

It has been estimated that between 25 and 80 percent of adults with CFS report a sudden illness onset, initiated by a flu-like illness or other acute infection. ".[18] In contrast, one population study that identified individuals meeting CFS criteria from a randomly selected population sample found that the majority of cases had a more gradual onset. Such discrepancies may derive from variations between the groups surveyed, as well as different interpretations of the meanings of "acute" and "gradual".[18]

Prospective studies have also examined the proportion of patients with acute, documented infections later develop CFS. 1 to 22 percent of patients with various documented acute infections (which included Epstein-Barr virus (EBV) associated glandular fever, non-EBV-associated glandular fever, Ross River virus, Giardia duodenalis enteritis, parvovirus B19, and Q fever) go on to develop CFS.[18]

--Wilshica (talk) 04:33, 12 October 2016 (UTC)Wilshica

Too soon?
I was tempted to add, under 'Research', a mention of an article headed "Metabolic features of chronic fatigue syndrome", published in PNAS. Then I noticed that it is quite recent (I just checked and it's the current edition of PNAS), so perhaps I should hold off till there are some published reactions? Any thoughts?

Article is at: http://www.pnas.org/content/113/37/E5472.short?rss=1&ssource=mfr Wayne 05:46, 24 October 2016 (UTC)

PACE trial content
Hi @Jfdwolff I am confused :- Why have you closed PACE trial tribunal ruling (QMUL v the IC and Matthees) and the PACE trial Update discussion. You say it is not a buliten board - both sections are for proposed changes. The PACE trial tribunal ruling (QMUL v the IC and Matthees) had no negative feedback on my proposed edit for 28 Days so consensus is implied and I will be making the edit in the near future. The PACE trial Update explicitly seeks to build a long term consensus on the subject so changes can be made. It appears to me that you are trying to close down the process of discussion and consensus building - this page covers a contentious issue and consensus building and discussion are important in ensuring it provides correct balanced information. You have also manually achieved Talk:Chronic_fatigue_syndrome_treatment which had a proposed edit (which referenced the NICE guidelines) which I asked for your help with. I have tried to follow your advice so please answer me so I can understand? I also welcome anyone else's point of view. C7762 (talk) 10:00, 28 September 2016 (UTC)
 * Can you restate or copy and paste your request or proposal in 1-2 sentences here again? I am having trouble following. Thanks.  Blue Rasberry   (talk)  13:28, 28 September 2016 (UTC)
 * Yes, . Please post an actual content proposal, with sources. Please do not use Talk pages in Wikipedia to provide "updates" on anything; that is not what they are for. Thanks. Jytdog (talk) 14:54, 28 September 2016 (UTC)

Thanks @Bluerasberry, There are two separate discussions that have been closed, both related to the PACE trial. The £5M trial has significant influence on national health treatment guidelines.
 * 1. With :-
 * Analysis (of PACE trial data according to original protocol) of Recovery rates for Graded Exercise Therapy and CBT reveals no statistical significance over specialist medical care. The Lancet article reported recovery rates of 22% for both GET and CBT - this was based on a mid-trial revised protocol where "13% of participants at baseline simultaneously met the trial eligibility criteria for ‘significant disability’ and the revised recovery criteria for normal self-reported physical function" . Where as analysis according to the original protocol reveals recovery rates of 7% (CBT) & 4%(GET) with no statistical significance over 3%(SMC)..
 * I was opening a debate about whether the level of emphasis on CBT & Graded Exercise Therapy and the underlying psychosocial model for CFS should be adjusted. This is not a proposed specific change but the opening of a bigger discussion which may take some time. While relevant to other CFS pages it is also relevant to parts of this page inc. Chronic_fatigue_syndrome sections. (Talk:Chronic_fatigue_syndrome is also relevant to this debate)


 * 2. I have posted a specific proposal for inclusion in the Chronic_fatigue_syndrome that had no objections for 28 days (this covers one of the most controversial issues in CFS and I put forward arguments why it should be included with no objections received) :-
 * Judgement in a First-tier Tribunal (20-22 April 2016) ordered the release of the PACE trial anonymised data under the Freedom of Information Act noting "there could be legitimate concerns that they wish to suppress criticism and proper scrutiny of their trial". .The Tribunal rejected the appellant Queen Mary University London claims that releasing data would breach patient confidentiality, might jeopardise follow up studies and damage the reputation of the college. . The tribunal found by a majority that the data “has been anonymised to the extent that the risk of identification is remote.”. PACE trial Researchers had the chance to present evidence use expert witnesses and be cross examined also argued that their was extremist activists in the CFS community, the Tribunal judged "assessment of activist behaviour was,in our view, grossly exaggerated and the only actual evidence was that an individual at a seminar had heckled Professor Chalder". Queen Mary University London issued a press statement stating "This has been a complex case and the Tribunal’s decision is lengthy. We are studying the decision carefully and considering our response, taking into account the interests of trial participants and the research community.". Alem Matthees the second respondent in the case is quoted as saying "This case ended up costing me greatly in time, energy, and health (currently bedridden) … but false or misleading claims of recovery or remission from debilitating illness simply have no place in the scientific literature.”
 * I also question why on page that requires discussion and reaching consensus which can take some time that it is now being auto archived after 3 weeks. I am looking for debate for or against my input not for it to be closed down. I am new to editing and welcome all constructive criticism to help me get it right. Thanks C7762 (talk) 15:02, 28 September 2016 (UTC)

Thanks @Jytdog, (please see above) - I can see that it might appear that my input about the analysis of the pace trial data was an update but it was an opening of a debate. If this is not appropriate then no problem but I would appreciate a post stating why there should be no debate rather than just closing it down. My specific proposed edit to the Chronic_fatigue_syndrome section has been there for 28 days with out any arguments against so I presume there are no objections. I am new to editing so all advice is welcome. Thanks C7762 (talk) 15:22, 28 September 2016 (UTC)

@Jytdog, I would also like to remind you that I originally posted the PACE trail Analysis under a discussion about Pathophysiology on this page (as I believed it was relevant) and it was yourself that sub sectioned it with the title Update. Thanks C7762 (talk) 15:49, 28 September 2016 (UTC)


 * Item #1 is not a proposal about changing article content so I will ignore it. Again, this Talk page is only for discussions about changing article content.  It is not a forum for general discussions about the topic.   About Item #2, as folks noted above this needs better sourcing and copyediting.  I have to do RW stuff but will come back to this tonight or tomorrow.  The same issues are being inserted into The Lancet article. argh. Jytdog (talk) 15:49, 28 September 2016 (UTC)


 * , You said, "I was opening a debate about whether the level of emphasis on CBT & Graded Exercise Therapy and the underlying psychosocial model for CFS should be adjusted. This is not a proposed specific change but the opening of a bigger discussion which may take some time"
 * There is room for some discussion but based on precedent I highly anticipate that you will continue to face struggle, frustration, and have discussions shut down. The Wikipedia community tends to have a quick publishing turnaround. If anyone is unable to make specific publishing suggestions quickly and concisely, then the time for publishing on Wikipedia for that topic may not be ripe.
 * You said, "I have posted a specific proposal for inclusion in the Chronic_fatigue_syndrome" At a glance, this information does not seem fundamental to a basic international understanding of CFS. This information seems to be about the PACE study. If you wanted to post this to an article about the PACE study, then that seems appropriate. If you insisted that this should go in the CFS article, then I would want to see more evidence that this information is universally relevant to understanding CFS. This looks like a regional lawsuit about a nuanced clinical trial without broadly confirmed significance.
 * Give more information if you have questions. Brevity matters a lot! Wikipedia does not have deep enough complexity to routinely need discussion. It is supposed to be easy to mirror content from cited sources.  Blue Rasberry   (talk)  20:07, 28 September 2016 (UTC)

Thanks @Bluerasberry, @Jytdog for your constructive input I take your points on board about discussions, edit suggestions and brevity e.t.c. I'm not going to push the PACE trial verdict (which was to be under Controversies section ) but I would say The PACE trial has had highly significant influence on the international perception of CFS and treatment guidelines in the UK,Europe, US, Austraila and further afield and the campaign to get the underlying data released was international including 42 international scientists and clinicians and the subsequent analysis. of the court ordered released data calls into question the published results and conclusions and associated treatment guidelines (CBT & GET). But obviously this is not time (and maybe not the place) so I will let it drop. Thanks C7762 (talk) 21:19, 28 September 2016 (UTC)


 * OK there are two levels of things here.
 * 1) One is about WP:Biomedical information and the questions are - "what was the study design of the PACE trial and what were its results? and  "Was the PACE trial valid"?  Sources for that need to comply with MEDRS
 * 2) The other is about how the PACE trial results have been used in creating guidelines, deciding insurance reimbursement, etc, and how people/organizations have reacted to that.  This is "Society and Culture" stuff, and sources need only comply with WP:RS.  We need to be careful in this part to not make statements about biomedical information.
 * These two levels are intermixed in the discussion above, and we need to keep them distinct.


 * I am going to start working on 1). MEDRS calls for literature review articles published in good journals, or statements by major medical/scientific bodies.  I did a pubmed search for "Chronic fatigue syndrome" AND PACE and got nothing relevant: here.  I did a pubmed search for "Chronic fatigue syndrome" AND CBT and got a slew of things, including Cochrane reviews: here.  I am going to start reading them to see what they have to say, and see if there is anything on point.  Will report back.  This will take time as there is a lot to wade through. Jytdog (talk) 21:54, 30 September 2016 (UTC)


 * Thanks for taking the time to look in to this. I would like to make the following input:-
 * 1. I suspect you will not find much CURRENTLY that criticises the PACE trial that complies with MEDRS. The data was only released for independent review a month or so ago.
 * 2. I think this complys with MEDRS :- In 2015 the National Institutes of Health that concluded that "continuing to use the Oxford definition may impair progress and cause harm. Thus, for needed progress to occur we recommend that the Oxford definition be retired" . The AHRQ then performed re-analysis omitting trials using the Oxford Definition (including PACE) and issued an addendum to their CFS advice severely downgrading the outcomes for CBT and Graded Exercise.
 * 3. Although I understand it will have no bearing on your review I would like to point out that a Principle Investigator in the PACE trial stated in court that the Cochrane review was not independent of the PACE trial as the 3 PACE PIs were on the board of the Cochrane review.
 * Thanks for listening.


 * C7762 (talk) 18:30, 20 October 2016 (UTC)
 * The NIH workshop report is not MEDRS.  I have been looking and not finding MEDRS sources so far; still looking. Jytdog (talk) 15:13, 21 October 2016 (UTC)
 * The Agency_for_Healthcare_Research_and_Quality systematic review and its 2016 addendum which summarizes research on methods of diagnosing and benefits and harms of multiple medical and nonmedical treatments and identifies evidence gaps should be compliant with MEDRS.


 * Thanks, C7762 (talk) 09:01, 30 October 2016 (UTC)

Thoughts on "Onset" section on the Chronic Fatigue syndrome page (reposted from archive)
The main review article that’s cited in this section is reliable (citation [18]), but the current phrasing is lifted word-for-word from that review, so needs to be either in quotes or rephrased. Also, there’s a final statement at the end, which is taken word-for-word from an earlier, poorer quality source (a conference abstract). I suggest omitting that.

I also suggest adding a line from the main review article already cited about the findings of prospective studies. This approach compensates partially for the problem of recall bias in retrospective studies.

So here’s a suggested rephrase:

It has been estimated that between 25 and 80 percent of adults with CFS report a sudden illness onset, initiated by a flu-like illness or other acute infection. ".[18] In contrast, one population study that identified individuals meeting CFS criteria from a randomly selected population sample found that the majority of cases had a more gradual onset. Such discrepancies may derive from variations between the groups surveyed, as well as different interpretations of the meanings of "acute" and "gradual".[18]

Prospective studies have also examined the proportion of patients with acute, documented infections later develop CFS. 1 to 22 percent of patients with various documented acute infections (which included Epstein-Barr virus (EBV) associated glandular fever, non-EBV-associated glandular fever, Ross River virus, Giardia duodenalis enteritis, parvovirus B19, and Q fever) go on to develop CFS.[18]

--Wilshica (talk) 04:33, 12 October 2016 (UTC)Wilshica — Preceding unsigned comment added by Wilshica (talk • contribs)

Endocrinological
I have replaced the text in this new section with the previous text about the HPA axis (which was deleted by IVI earlier this year). There was some discussion about restoring this section at the time, but it never happened. --sciencewatcher (talk) 17:13, 16 November 2016 (UTC)


 * I see Doc James replaced "there is a substantial body of evidence" with "there may be". That doesn't seem to tally with the review, which says "a substantial body of evidence now points to the following four main findings in patients with this condition: a degree of mild hypocortisolism; attenuated diurnal variation of cortisol; enhanced negative feedback; and blunted HPA axis responsiveness". --sciencewatcher (talk) 02:11, 17 November 2016 (UTC)
 * Um so it is a copyright infringement than?
 * We do not use the term patient generally and we must paraphrase as that is Elsevir. Doc James  (talk · contribs · email) 03:42, 18 November 2016 (UTC)
 * Yikes. Sciencewater please don't restore the copyvio. Jytdog (talk) 03:53, 18 November 2016 (UTC)


 * I'm pretty sure copying 4 words is not a copy violation. In fact it would generally be preferred to copy the level of evidence verbatim (e.g. "moderately effective") rather than trying to use a different word which may not be the same. But Doc James' edit seems fine in this case. --sciencewatcher (talk) 15:19, 18 November 2016 (UTC)

Mitochondrial dysfunction?
Is this source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2680051/ A reliable source? Everything seems to check out but something doesn't feel right. The journal "International Journal of Clinical and Experimental Medicine" has been controversial, and research for "ATP profile tests" don't come up with much. — Preceding unsigned comment added by Petergstrom (talk • contribs) 07:14, 18 November 2016 (UTC)


 * Given the recent study that shows higher than normal ATP in patients, it seems to contradict this study. That is a good example why we always use reviews where possible (see MEDRS). --sciencewatcher (talk) 15:20, 18 November 2016 (UTC)

Could you send that study to me?Petergstrom (talk) 16:11, 18 November 2016 (UTC)


 * I don't have it. Have a look on the reddit/healthrising/phoenixrising. I'm pretty sure I saw it there (probably reddit). Or look at recent research on google scholar. --sciencewatcher (talk) 16:42, 18 November 2016 (UTC)
 * Lead author is a controversial GP whose "science" in previous papers hasn't been, shall we say, rigorous. I would suggest a careful assessment of any paper authored by her. She has a wiki for patients advocating all sorts of dubious medicine. I'm not surprised that this study has been contradicted. It happens a lot! -Roxy the dog bark 11:51, 20 November 2016 (UTC)
 * More. Second author is the chief of the lab that performs the "ATP profile tests" that cost a considerable sum, as they cannot be done on the NHS, probably because they aren't accepted as telling anything significant. I don't think there is another lab in the world that does these tests, which were, I believe, invented by said second author. -Roxy the dog bark 13:44, 20 November 2016 (UTC)

Some of the Challenges to the PACE Trial
The article currently claims that:


 * "Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases.[8] Exercise does not make people worse.[9]"

The former statement is doubtful, at best; I am not an expert, but have read that the main evidence for graded exercise therapy has been the PACE trial, and the methods and interpretation have been challenged. The latter statement is false as written, since chronic fatigue syndrome often involves post-exertional malaise. (See the symptom list.) 96.255.9.115 (talk) 04:32, 21 November 2016 (UTC)

Some of the challenges, this year, include:
 * Frank Twist, "PACE: CBT and GET are not rehabilitative therapies"
 * Sten Helmfrid, "Studies on Cognitive Behavioral Therapy and Graded Exercise Therapy for ME/CFS are misleading"
 * James Charles Coyne and Keith R. Laws, " Results of the PACE follow-up study are uninterpretable"
 * Mark Vink, "The PACE Trial Invalidates the Use of Cognitive Behavioral and Graded Exercise Therapy in Myalgic Encephalomyelitis/ Chronic Fatigue Syndrome: A Review"
 * The first ref is a letter to the editor, not a useful ref. The 2nd is published in a journal that is not indexed by pubmed or medline.  Not useful.   The third is another letter to the editor. Not useful. The last was published in a journal run by Sci Forschen, which is on Beall's list of predatory publishers.  I was hopeful on that one, but it too is not useful. Jytdog (talk) 05:22, 21 November 2016 (UTC)


 * If you take a quick look at the edit history for this article, you'll see the info on CBT and GET was mostly written before PACE came out. Although we mention PACE, the evidence is mostly taken from reviews that cite a number of studies (and I'm not sure if they even include PACE). So, no, the main evidence for GET (or CBT) is not from the PACE trial.--sciencewatcher (talk) 06:33, 21 November 2016 (UTC)
 * What we currently have is supported by much better references than those criticizing it. Doc James  (talk · contribs · email) 12:00, 22 November 2016 (UTC)


 * Think the OP has a very good point, CFS is umbrella  term. Sub groups of suffers do get worse when under going graded exercise. Many a doctor, is ignorant of the the many aetiologies,   thus lumps it into Oh you just have CFS.  WP is an encyclopedia in progress and there is plenty of research showing that in some sub groups, their mitochondria disfunction makes a program of  graded exercise contra indicated. Ie. When the patents progress regresses,  it is not the patents fault but inappropriate treatment. This was known and acknowledged since the times of Galen. Even then, doctors were still  bleeding their  patents to death because they were taught that it was the right thing to do, despite the evidence in front of their very eyes. Orthodoxy (definition: right thinking) is one thing but wontent  ignorance (to wilfully ignore, due to knowing one is always right) and thus cherry pick  to suit ones own received teachings on WP is another.--Aspro (talk) 18:37, 22 November 2016 (UTC)


 * The problem is that there is no reliable source showing this. There is also no evidence that some subgroups have mitochondrial dysfunction (recent research seems to have debunked that, as noted above). --sciencewatcher (talk) 18:47, 22 November 2016 (UTC)


 * The citation for the claim "Exercise does not make people worse." Is invalid. The Cochrane review states "no evidence suggests that exercise therapy may worsen outcomes" - this is not the same as exercise in general. Secondly, none of the articles reviewed by the Cochrane study used objective measures of activity/exercise. We literally have no evidence whether patients actually increased their activity levels, or simply substituted one type of activity (daily chores etc) for another. See: Does graded activity increase activity? A case study of chronic fatigue syndrome. Friedburg, Journal of Behavior Therapy and Experimental Psychiatry 33(3-4):203-15 · September 2002.
 * Secondly, the quality of harms reporting in clinical trials of graded exercise therapy has been reported as poor and does not meet the same standards of contemporary drug trials. Lastly, short term harms of exercise have been consistently reported, "Influence of exhaustive treadmill exercise on cognitive functioning in chronic fatigue syndrome" "Increased daily physical activity and fatigue symptoms in chronic fatigue syndrome" "Demonstration of delayed recovery from fatiguing exercise in chronic fatigue syndrome" etc.
 * Therefore The claim "Exercise does not make people worse" is not consistent with the current published body of evidence and needs to be removed from the article.

27.33.143.13 (talk) 01:43, 23 November 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 one external links on Chronic fatigue syndrome. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20110718151623/http://www.iacfsme.org:80/PACETrial/tabid/450/Default.aspx to http://www.iacfsme.org/PACETrial/tabid/450/Default.aspx
 * Added archive https://web.archive.org/web/20110625051513/http://cfids.org/pdf/lancet-analysis.pdf to http://www.cfids.org/pdf/lancet-analysis.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 07:22, 24 November 2016 (UTC)

"Onset" section on the Chronic Fatigue syndrome: Some thoughts
I am bumping this from the archived talk pages - some editors would like to discuss further before any change is implemented.

The main review article that’s cited in this section is reliable (citation [18]), but the current phrasing is lifted word-for-word from that review, so needs to be either in quotes or rephrased. Also, there’s a final statement at the end, which is taken word-for-word from an earlier, poorer quality source (a conference abstract). I suggest omitting that.

I also suggest adding a line from the main review article already cited about the findings of prospective studies. This approach compensates partially for the problem of recall bias in retrospective studies.

So here’s a suggested rephrase:

It has been estimated that between 25 and 80 percent of adults with CFS report a sudden illness onset, initiated by a flu-like illness or other acute infection. ".[18] In contrast, one population study that identified individuals meeting CFS criteria from a randomly selected population sample found that the majority of cases had a more gradual onset. Such discrepancies may derive from variations between the groups surveyed, as well as different interpretations of the meanings of "acute" and "gradual".[18]

Prospective studies have also examined the proportion of patients with acute, documented infections later develop CFS. 1 to 22 percent of patients with various documented acute infections (which included Epstein-Barr virus (EBV) associated glandular fever, non-EBV-associated glandular fever, Ross River virus, Giardia duodenalis enteritis, parvovirus B19, and Q fever) go on to develop CFS.[18] --Wilshica (talk) 05:07, 25 November 2016 (UTC)

Psychological subhead
The section on psychological factors has a very different tone to the other sections on possible causes. These other sections cite evidence for their respective causal factors. This section states outright: "Current data suggests that genetic, physiological and psychological factors work together to precipitate and perpetuate the illness. " This is a speculation that goes well beyond the data. It is not yet clear on the current evidence that all of these factors "work together to precipitate and perpetuate the illness".

This section should cite any relevant evidence supporting psychological causal factors in the development and/or maintenance of CFS, without making any sweeping assertions as to causation.

Do others agree?

--Wilshica (talk) 23:30, 21 November 2016 (UTC)wilshica


 * Well, that is what the Afari review says, and it looks like that statement is based on good evidence. Are you suggesting that the psychological section should be expanded to better explain the possible factors? --sciencewatcher (talk) 01:20, 22 November 2016 (UTC)


 * No. I'm saying we stick to evidence in these sections, no unsubstantiated claims (someone once making that claim in a publication does not make it WP-worthy). All the other sections on causation stick to the evidence. I think that's what we want to do here. I don't think there's any evidence to support the claim that "genetic, physiological and psychological factors work together to precipitate and perpetuate the illness". At best, its a rather empty statement; at worst, its misleading - we don't really know whether these factors work together or operate independently, or even which ones are most important at this stage. --Wilshica (talk) 08:32, 22 November 2016 (UTC)Wilshica
 * We should get rid of the ref from 2003 and use more contemporary sources more, is what we should do. The IOC and CDC don't talk about psyc causes at all. Jytdog (talk) 10:11, 22 November 2016 (UTC)


 * What about https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2292451/ or http://www.nature.com/nrn/journal/v12/n9/full/nrn3087.html or https://www.researchgate.net/publication/7083503_Chronic_fatigue_syndrome_a_systematic_review? --sciencewatcher (talk) 17:30, 22 November 2016 (UTC)
 * Those are also all older than the 2015 IOM report (which reviewed everything to date) and the CDC criteria.  The NHS Choices site which we also already cite is very reliable and is also more recent (2015) - it does mention possible psychiatric causes.  So we do have one very good recent ref for psychiatric (not psychological) Jytdog (talk) 18:48, 22 November 2016 (UTC)


 * The systematic review Sciencewatcher refers to is probably not as up to date as we'd like (from 2006). The Nature ref is interesting because it represents multiple viewpoints. But it is fives years old too. The IOM report would seem to be a good tertiary ref here, and good to include something from outside the UK. The NHS choices webpage just says that many diverse theories have been proposed, and lists them all; it doesn't review any evidence or support any particular causal model. Its not much use to us here. --Wilshica (talk) 01:36, 23 November 2016 (UTC)Wilshica

Given the uncertainty and controversy in this field, we do need to keep evidence and claims separate. What we need to do here is describe the science, not the theories (there are other sections for that). — Preceding unsigned comment added by Wilshica (talk • contribs) 01:30, 23 November 2016 (UTC)

How about this for the "Psychological" subhead? It sticks to the evidence, and doesn't make any sweeping claims (the last version also made an implicit assumption that correlation=causation, I have made the causal reasoning explicit):


 * The disease profile in chronic fatigue syndrome is distinguishable from primary depression. (cite Zarouf ref here from Sciencewatcher). However, secondary anxiety and depression are common in patients with CFS.(cite Zarrouf ref again here) It is not yet clear whether anxiety, depression and other psychological factors are a cause or a consequence of the physical disability in CFS. Patients with more severe physical symptoms are also more likely to view their symptoms as having a physical basis.[5] It is possible that such attributions may play a causal role in maintaining the disease; however, it may be the case that persons with more severe symptoms are more inclined to view them as physical. Uncertainties about the causes, prognosis, diagnosis and treatment, and the lack of clinical guidance for healthcare professionals, aggravate the impact of the disease both in those affected and their caregivers.[45]--Wilshica (talk) 06:10, 25 November 2016 (UTC)

Activity of the BG in CFS patients
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0098156

Although this is not a secondary source, decreased BG activity has been found in patients without pathological fatigue. Is this something that can be added? Petergstrom (talk) 21:09, 1 December 2016 (UTC)
 * We should wait for a secondary source IMO. Doc James  (talk · contribs · email) 23:50, 5 December 2016 (UTC)


 * Multiple MRI scans by different independent researches have noted this phenomena for something like a decade. Bankhead is a Writer, so this is a secondary source: . A quick google shows there are many more. So one doesn’t need to wait anymore. It's accepted by neurologists (that interpret the scans) but not by some of the other medical establishment that are not up-to speed and that is probably where the problem lays. --Aspro (talk) 01:21, 8 December 2016 (UTC)
 * Medpage today is not MEDRS either. Jytdog (talk) 02:22, 8 December 2016 (UTC)

These are all the sources I could find. https://www.ncbi.nlm.nih.gov/pubmed/24858857, http://www.nature.com/articles/srep21386, http://www.fasebj.org/content/26/1_Supplement/1035.20, http://bmcneurol.biomedcentral.com/articles/10.1186/1471-2377-4-14. The second one is on general fatigue not CFS. Petergstrom (talk) 03:12, 8 December 2016 (UTC)
 * right, so if something isn't discussed in reliable sources (and MEDRS defines reliable sources) then it doesn't get any weight in WP; we don't discuss it. Jytdog (talk) 03:15, 8 December 2016 (UTC)


 * Indeed. No high quality secondary source = no mention. Especially in this article. JFW &#124; T@lk  11:00, 8 December 2016 (UTC)

new content & source on HPA/immune interplay
The following was added here on Dec 5; contested today due to the source here and here, and i removed it here and want to discuss it. Here it is:

The HPA axis dysfunction may be due to oxidative and nitrosative stress and inflammation

First the source is fine, a recent review in a high quality journal. What I am struggling with is the content. The conclusion of this paper is that "The stress crash theory that HPA axis hypofunction drives the activation of immune-inflammatory pathways in ME/CFS cannot be validated. This theory is contradicted by a number of findings, including (a) HPA axis hypofunction is only present in a subgroup of patients, (b) there are no changes in HPA axis during the early stages of ME/CFS, and (c) the immunosuppressive effects of GCs are enhanced in ME/CFS." and it goes on from there.

User: Petergstrom we have talked over-simplification in the past, and it is not valid - based on this source you brought, as well as a bunch of others - that there are no super-clear diagnostic criteria or common signs across everyone with the CFS. This is messy and you cannot write stuff like "The HPA axis dysfunction may be due to...." as though HPA hypofunction is a definitive hallmark of the condition as it is currently understood. I am going to add this back with nuance. Jytdog (talk) 01:26, 11 December 2016 (UTC)


 * The Papadopoulos source is higher quality and has less speculation. It says "A multidimensional etiological model remains most probable, with low cortisol levels occurring at a later stage of the illness, moderated by factors such as activity levels, depression, early-life stress and psychotropic medication". --sciencewatcher (talk) 01:47, 11 December 2016 (UTC)
 * that is five years old, right on the edge of being too old per WP:MEDDATE and it is four years before the major NAS review which is still dramatically under-cited in this article. Jytdog (talk) 01:57, 11 December 2016 (UTC)
 * Another review is https://www.hindawi.com/journals/isrn/2013/784520/ --sciencewatcher (talk) 02:57, 11 December 2016 (UTC)

Should the opening sections on this page include the statement "Exercise does not make people worse"?
I am raising this comment from an unidentified WP editor to its own seciton

The citation for the claim "Exercise does not make people worse." Is invalid. The Cochrane review states "no evidence suggests that exercise therapy may worsen outcomes" - this is not the same as exercise in general. Secondly, none of the articles reviewed by the Cochrane study used objective measures of activity/exercise. We literally have no evidence whether patients actually increased their activity levels, or simply substituted one type of activity (daily chores etc) for another. See: Does graded activity increase activity? A case study of chronic fatigue syndrome. Friedburg, Journal of Behavior Therapy and Experimental Psychiatry 33(3-4):203-15 · September 2002. Secondly, the quality of harms reporting in clinical trials of graded exercise therapy has been reported as poor and does not meet the same standards of contemporary drug trials. Lastly, short term harms of exercise have been consistently reported, "Influence of exhaustive treadmill exercise on cognitive functioning in chronic fatigue syndrome" "Increased daily physical activity and fatigue symptoms in chronic fatigue syndrome" "Demonstration of delayed recovery from fatiguing exercise in chronic fatigue syndrome" etc. Therefore The claim "Exercise does not make people worse" is not consistent with the current published body of evidence and needs to be removed from the article.

27.33.143.13 (talk) 01:43, 23 November 2016 (UTC)

Its perhaps not fair to say that this claim is invalid (it has some empirical basis), but I do agree that we should be cautious more when commenting on possible adverse/side effects of treatments. And behavioural treatments should be no exception. As the editor above noted, exercise physiology studies show that some patients can have an adverse reaction to intense exertion, at least over the short term. Informal patients surveys also report a high incidence of adverse effects from exercise programmes in general (and these sorts of reports are part of any consideration of treatment adverse/side effects). A more cautious phrasing would be preferred. The suggested edit looks acceptable to me.

--Wilshica (talk) 02:06, 23 November 2016 (UTC)


 * The blanket claim "Exercise does not make people worse." is invalid. The cited reference stated that there were no major harms in the context of supervised exercise therapy of those who agreed to participate in a clinical trial (eg there is selection bias involved). This is very different from the Carte blanche claim that exercise is not harmful. It is not ethical for a doctor, nor this article to make such a broad claim given there is evidence of harm of exercise in other studies. Not only exercise physiology studies, but simple exercise programme studies. eg Black et al 2005 (DOI: 10.1186/1476-5918-4-3) Arch 14:37, 24 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * Sure changed to "exercise therapy"
 * We do not use a more than 10 year old primary source to refute a recent high quality secondary source. Doc James  (talk · contribs · email) 03:06, 25 November 2016 (UTC)
 * The statement "Exercise therapy does not make people worse." is still a non-sequitur based on the cited reference. First of all, the Cochrane review only considered "serious averse effects" not "making people worse" in general. The Cochrane review specifically stated "Serious adverse reactions were rare in both groups, but sparse data made it impossible for review authors to draw conclusions." Only two studies were considered when reviewing serious adverse effects and the quality of harms reporting in those two studies has been questioned compared to the quality expected of drug trials, see the feedback section of the Cochrane review. http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003200.pub5/full
 * What part about "sparse data made it impossible for review authors to draw conclusions" leads you to think this is high quality evidence? Arch 03:44, 25 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * It states "no evidence suggests that exercise therapy may worsen outcomes" so what we have is a good enough summary. Doc James  (talk · contribs · email) 06:38, 25 November 2016 (UTC)
 * No, what you have added is your own point of view, not a fair summary of the evidence presented. The article explicitly states that it is impossible to draw conclusions due to sparse data. If insufficient evidence has been collected/reviewed, then you cannot conclude the converse that "exercise therapy does not make appear to make people worse".Arch 12:31, 25 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * That is not Doc James POV, but from the authors conclusion of the review, which you probably haven't read. it states " ... no evidence suggests that exercise therapy may worsen outcomes" Which part of that don't you understand. Seems pretty clear, dont you think. -Roxy the dog. bark 14:09, 25 November 2016 (UTC)
 * The context was that they stated "sparse data made it impossible for review authors to draw conclusions" in the review. A lack of evidence (due to sparse data) for efficacy of a treatment is not the same as saying that treatment is effective. Likewise, a lack of evidence for adverse effects (due to sparse data) of a treatment is not the same as saying that treatment is safe.Arch 15:20, 25 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * By the way, if this sparse data (when cited in systematic reviews) is acceptable, then using this logic one could conclude that Hydrocortisone, Ritixumab "appear to be effective" (they have shown benefit in blinded randomised trials) according to "A Systematic Review of Drug Therapies for Chronic Fatigue Syndrome/Myalgic Encephalomyelitis." doi: 10.1016/j.clinthera.2016.04.038. But I would not add such statements because the evidence is currently lacking. Arch 15:38, 25 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)


 * Have removed statement “Exercise therapy does not make appear to make people worse” for the obvious medical philosophical reason that medical scientific reason  conforms and upholds  to well accepted scientific methodology.  CFS  has become an umbrella term for more than one aetiology (the whole point in creating the International Classification of Diseases was so that each disease had both a definition and best proven treatment protocol). If some studies select a cohort of subjects according to one definition of CFS and they appear to  benefit from an exercise regime that does not (using the scientific method) equates in any form nor medical sense, that all CFS  subjects benefit from  an exercise regime without suffering harm. Thus,  it can't be stated that Exercise therapy does not make appear to make people worse If any editor wants to revert  that, then may they kindly state a Cochrane  study or some such, that states  that one can just pick and choose cohorts to prove a mere hypothesis. Some editors here appear to have free access to med libraries and belong to medical  orgs that provide them with free access to med journal and databases. They should not  be pontificating  the dogma they were taught in med school but using their learning during  their long hours of studies ( unless the went to parties every night like a lot of med students did) an expounding  the latest  up to date encyclopedic information. Cross examined of course  by the knowledge they gained by declining invitation to parties when they knew they should be studying  --Aspro (talk) 14:40, 25 November 2016 (UTC)
 * We follow high quality secondary sources. Doc James  (talk · contribs · email) 22:59, 25 November 2016 (UTC)
 * When the secondary source explicitly states that there is an ABSENCE OF EVIDENCE for a claim "but sparse data made it impossible for review authors to draw conclusions", you cannot use this to make a positive claim about the safety of that treatment. The addition "Exercise therapy does not appear to make people worse" is POV statement that is not justified by the secondary source. Arch 01:27, 26 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * The content added is supported by the reference in question. Doc James  (talk · contribs · email) 01:33, 26 November 2016 (UTC)
 * Doc James, I'm asking you a specific question: What do you think "no evidence" and "but sparse data made it impossible for review authors to draw conclusions" means in context? If I cited a secondary study that said "sparse data made it impossible for review authors to draw conclusions" in reviewing the efficacy of a treatment, would you let that edit stand? Arch 01:38, 26 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)


 * Doc James, we have discussed your proposed edit on this talk page as requested (adding "exercise does not make people worse"), and various editors have given their views and do not think it appropriate. I think this would be a good time to stop your attempts to reinsert this statement each time it is deleted, and focus on other WP material that needs improvement.
 * --Wilshica (talk) 03:49, 26 November 2016 (UTC)
 * Okay will start a RfC. The removal of a very recent Cochrane review is concerning. Doc James  (talk · contribs · email) 04:54, 26 November 2016 (UTC)

RfC: Suggested addition to the lead
We have a June 2016 Cochrane review that states "Patients with CFS may generally benefit and feel less fatigued following exercise therapy, and no evidence suggests that exercise therapy may worsen outcomes."

As this is an important point I propose we add "Exercise therapy does not appear to make people worse." to the third paragraph of the lead.
 * Option 1 (proposed by Doc James)

Other wording could include "There is no evidence of harm from exercise." Doc James (talk · contribs · email) 05:01, 26 November 2016 (UTC)
 * Option 2 (proposed by Doc James)

"In a systematic review of exercise therapy, no evidence of serious adverse effects was found, however data was insufficient to form a conclusion." The Cochrane review explicitly stated they could not form a conclusion on adverse effects, so a point along those lines needs to be stated. Arch (talk) 04:28, 27 November 2016 (UTC)
 * Option 3 (proposed by Architectonic)


 * Option 4 (proposed by RexxS}
 * "There is no evidence of worsened outcomes from exercise therapy compared with other interventions, but data on serious adverse effects is insufficient to form any conclusions." The Cochrane review explicitly stated they could not form a conclusion on serious adverse effects, but the authors were able to reach a conclusion on worsening outcomes in general: no evidence for it.

Support option 1

 * Support as proposer. Cochrane reviews are high quality sources and it was recently published. Doc James  (talk · contribs · email) 05:01, 26 November 2016 (UTC)
 * support it is a new review and is from Cochrane reviews (usually they are the best...IMO)--Ozzie10aaaa (talk) 11:44, 26 November 2016 (UTC)
 * Support: the best quality evidence we have on "the effects of exercise therapy for patients with CFS as compared with any other intervention or control" is from the Cochrane review. The authors' conclusions begin with the sentence "Patients with CFS may generally benefit and feel less fatigued following exercise therapy, and no evidence suggests that exercise therapy may worsen outcomes." This conclusion is relevant to the subject and is not contradicted by another equivalent secondary source. There can be no doubt that the proposed text should be included in the article. --RexxS (talk) 22:29, 26 November 2016 (UTC)
 * Support - Cochrane reviews are top sources and they tend to get more weight. Option 1 adds better detailed content for the reader. Option 2 is shorter and okay. QuackGuru  ( talk ) 23:54, 26 November 2016 (UTC)


 * Weak support - think not much worth a mention, but if you've got to have one, this one seems closest to summarizing the cite with WP:INTEGRITY. Mostly I noted that Cochrane was not showing effective for treating CFS, with the likely benefits mentioned are in items of sleep, health, mental self-view.  The other options above I toss mostly as they improperly claim "no evidence" [of harm] but Cochrane says "Serious adverse reactions were rare"..."sparse data made it impossible for review authors to draw conclusions."  Sounds to me like folks with CFS who exercise become athletic CFS people except a few hurt themselves exercising not really a treatment for CFS.  Markbassett (talk) 02:07, 28 November 2016 (UTC)
 * Support. This seems the best way to summarise the findings. 2, 3 and 4 are also good. Maproom (talk) 09:22, 28 November 2016 (UTC)
 * Support. Best of the three viable proposals. Option 3 is not supported by the review. -Roxy the dog. bark 10:48, 28 November 2016 (UTC)

Support option 2

 * Support happy with this version aswell. Doc James  (talk · contribs · email) 06:58, 26 November 2016 (UTC)
 * Support. Not the best option for me, but nevertheless. -Roxy the dog. bark 10:51, 28 November 2016 (UTC)
 * Support.Per WP:lead the lead contains a summary of the article, in non-technical wording, as possible. The article body contains the detailed explanation.CuriousMind01 (talk) 13:09, 1 December 2016 (UTC)

Support option 4

 * Support as reasonable aswell. Doc James  (talk · contribs · email) 05:36, 27 November 2016 (UTC)
 * Support. This is the best one.-- Literaturegeek | T@1k? 07:06, 27 November 2016 (UTC)
 * Support. OK too. -Roxy the dog. bark 10:53, 28 November 2016 (UTC)
 * Support This appears to be the only reasonable summary of the cochrane review, everything else changes the tenor of their statements. SPACKlick (talk) 13:41, 28 November 2016 (UTC)
 * Support My first preference is no addition (discussed below). But as a compromise, I'm willing to support option 4 as it is a fair representation of what is stated in the study (unlike options 1 & 2). Arch (talk) 05:50, 29 November 2016 (UTC)
 * Support, most balanced presentation of the conclusions. JFW &#124; T@lk  16:28, 30 November 2016 (UTC)
 * Support Most reasonable summary.  Nik ol ai Ho ☎️ 04:32, 2 December 2016 (UTC)
 * Support Doesn't quite roll of the tongue but it's the most accurate. Iazyges   Consermonor   Opus meum  04:43, 2 December 2016 (UTC)
 * Support. After reading the below discussion, this one seems to be the most accurate representation of the material from the linked source. Lizzius (talk) 20:18, 7 December 2016 (UTC)

Oppose

 * The statement "Exercise therapy does not appear to make people worse." takes the quote "no evidence suggests that exercise therapy may worsen outcomes." out of context and does not reflect the evidence provided in the article that was cited.


 * ·The cited Cochrane review only reviewed a single study (compared to "‘Passive control’ (treatment as usual/waiting-list control/relaxation/flexibility") and simply re-stated the results without meta-analysis. The authors conclusion that there was "no evidence" is to be taken literally, that there is an absence of evidence in studies that met the review criteria. The meaning of this statement is not ambiguous as the authors explicitly clarified in the article that "sparse data made it impossible for review authors to draw conclusions." specifically about adverse effects. The fact that the authors stated that it was impossible to draw conclusions means the proposed statement "Exercise therapy does not appear to make people worse." reflects a personal point of view and does not reflect the evidence in the cited article.


 * ·This review only reported "serious adverse effects", this is much more specific than "make people worse", which could involve any type of temporary or long term adverse effects as a result of a treatment. The statement "make people worse" is far too non-specific. Arch 05:44, 26 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * Per the paper "We have included eight randomised controlled studies and have reported data from 1518 participants in this review." Doc James  (talk · contribs · email) 06:14, 26 November 2016 (UTC)
 * Yes and if you look closely under the table "Comparison 1. Exercise therapy versus treatment as usual, relaxation or flexibility", you will see only one study was included in the adverse events analysis compared to "treatment as usual/waiting-list control/relaxation/flexibility". The effect size was stated as "Totals not selected" in the table. Here is the direct link to the figure: http://onlinelibrary.wiley.com/enhanced/figures/doi/10.1002/14651858.CD003200.pub5#figure-viewer-CD003200-fig-00103 Arch 06:30, 26 November 2016 (UTC) — Preceding unsigned comment added by Architectonic (talk • contribs)
 * Sure but they looked for evidence of harm in all the avaliable RCTs. Doc James  (talk · contribs · email) 06:33, 26 November 2016 (UTC)
 * Yes and the other RCTs did not meet the quality criteria for inclusion in the adverse effect analysis. Anyway, the unavoidable bottom line is that the article clearly states "sparse data made it impossible for review authors to draw conclusions." on the point of adverse effects. To conclude in the Wikipedia article is inserting your own point of view. Arch (talk) 06:49, 26 November 2016 (UTC)
 * No, the sparse data made it impossible for the authors to draw conclusions on "serious adverse reactions". They clearly saw enough data concerning all adverse reactions to conclude that "no evidence suggests that exercise therapy may worsen outcomes", because that's exactly what they concluded. That conclusion carries the weight not only of the authors, but of the Cochrane review. Your bit of amateur analysis carries no weight whatsoever. --RexxS (talk) 22:40, 26 November 2016 (UTC)

Option 2 lacks clarity. It is dishonest to imply something is safe due to "no evidence", when that means lack of evidence. It would be much more accurate to say "Evidence is currently lacking on whether exercise (or exercise therapy) leads to harm". Arch (talk) 07:06, 26 November 2016 (UTC)
 * As such, I oppose both option 1 and option 2. I may consider other options. Arch (talk) 09:06, 26 November 2016 (UTC)


 * For reasons outlined above and below (see discussion subsection), I oppose both current options. Option 2 is even worse than Option 1. Alternatives should be focused on caveats with the evidence rather than presenting poorly evidenced statements as fact while ignoring the context provided in the same source. Perhaps "Patient surveys consistently report harms from exercise therapy, but a general lack of data from randomised trials makes it difficult to draw conclusions about the safety of exercise therapy for CFS." -- Tekaphor  ( TALK ) 11:54, 26 November 2016 (UTC)


 * (just adding my discussion points from below in here under "oppose" for clarity) My feeling is that proper explanation and qualification is needed, and all that would suggest a separate, later section, perhaps one dedicated to graded exercise therapy. This is an opening paragraph. I would rather not see treatments discussed at all at this stage. But if they are the statement needs to be very brief: the presence/absence of adverse effects of a treatment requires care and caution. I also do not want to encourage patients to go out and exercise madly, that's reckless. The statement in the Cochrane review is meant to be interpreted within the context of certain specific supervised programmes, and even then they never claim there is "no risk". I am unclear why we are still discussing this, as the overwhelming consensus is that the statement should not appear.--Wilshica (talk) 07:27, 26 November 2016 (UTC) --Wilshica (talk) 23:40, 26 November 2016 (UTC)

Strongly oppose: Cochrane reviews may be of high quality, yet one has to judge and place them into context to gain any benefit from them. The review is of studies where the cohort had be selected from the umbrella term that covers CFS in all its manifestations. Not to put to finer point on it, five out of those eight used the used the Oxford Criteria. Hardly a random CFS cohort! Therefore, to promote these inclusions in this article (as if it was a matter of fact) is a bit like saying that  “ A Cochrane review found that a few  patients selected for a certain type of cancer benefited from a ce'rtain type of chemotherapy and thus all cancer patents will suffer no harm  at all from this type of treatment”. The supporting review provided - even states -  that "Serious adverse reactions were rare” I.e. not totally absent. Although some editors may have enough medical degrees to get a job as a clinical thermometer they appear to be ignorant of the science concerning aetiology and the work that gets done  to find the best treatments.  So the sentence proposed can not stand as worthy of inclusion in this article. --Aspro (talk) 12:37, 26 November 2016 (UTC)


 * To further the comparison of exercise for CFS to chemotherapy for cancer, the evidence shows harms are not recorded and patients are not receiving chemotherapy, yet blanket statements are still being made about its safety. In clinical trials, patients are told that they need it and how beneficial it is, instructed on how receive it if they want it, and a small proportion even report benefit from being in the programme, but there is no objective evidence of the chemotherapy being regularly administered (i.e. no objective evidence that CFS patients are sustaining gradual increases in activity, actually the evidence shows no such increases). Meanwhile thousands of patients outside the clinical trials are reporting the 'chemotherapy' (exercise) makes them worse but these are dismissed as not "MEDRS". The current options are dangerous and misleading. -- Tekaphor  ( TALK ) 13:33, 26 November 2016 (UTC)


 * Aspro has a good point about the safety issue being further complicated by the use of the Oxford criteria to select CFS patients and then making blanket statements about safety that may not apply to other patients. US health authorities have called for the Oxford criteria to be retired because "using the Oxford case definition results in a high risk of including patients who may have an alternate fatiguing illness or whose illness resolves spontaneously with time" and "continued use of the Oxford (Sharpe, 1991) case definition may impair progress and cause harm" . -- Tekaphor  ( TALK ) 13:33, 26 November 2016 (UTC)
 * No, forming your own judgements about the content of a reliable secondary source is precisely what WP:MEDRS says not to do: "Do not reject a high-quality type of study (e.g., a meta-analysis) ... because of personal objections to the inclusion criteria, references, funding sources, or conclusions in the higher-level source." We can make judgements about what type of review it is, but Wikipedia editors are not qualified to critique the details of a systematic review to advance their own point-of-view as you're doing. Criticism of a secondary source is valid when published in an equally authoritative publication, but not just on the say-so of a Wikipedia editor. If you have found an authoritative published refutation of the Cochrane review, then by all means, refer to it. Otherwise, your self-produced analysis carries no weight here. --RexxS (talk) 22:59, 26 November 2016 (UTC)


 * Oppose the lead already states "The fatigue of CFS is not due to ongoing exertion, is not much relieved by rest, and is not due to any other medical condition." ( not sure if this was inserted post discussion), no need to state this twice--Tom (LT) (talk) 21:53, 26 November 2016 (UTC)
 * They are different statements. One is looking at the cause the other is looking at the side effects of a treatment. Doc James  (talk · contribs · email) 23:09, 26 November 2016 (UTC)

Oppose either misleading summary - I think it's important people read the review before deciding on this, including the comments from Tom Kindlon and Robert Courtney, and the author's response. Ideally people would be aware of the public exchange between James Coyne and David Tovey too. Other than with the PACE trial, reporting of harms has been recognised as being generally poor for exercise trials of CFS: https://www.ncbi.nlm.nih.gov/pubmed/26112761 Considering the current controversy surrounding the PACE trial (the studies authors recently losing a court case in which they had argued that Cochrane's review of their data was not independent as they themselves had been involved in the process, so still had control) it seems odd to place such a bold statement founded on this trial so prominently, eg: https://www.statnews.com/2016/09/21/chronic-fatigue-syndrome-pace-trial The Cochrane review only looked at outcomes for RCTs, so any summary of its findings must recognise this, particularly given the negative reports seen from patients receiving exercise therapy and the research seeming to indicate abnormal responses to exercise. A lack of evidence of harm from one type of evidence is not the same as there being no evidence of harm. I don't know much about wikipedia, and was told about discussion by someone using it as an example of how wikipedia's rules seem to actively discourage critical thinking. This Cochrane review classed the PACE trial as having a low risk of selective reporting - a judgement by the authors that always seemed to defy Cochrane's own guidelines, and one that has now been left looking absurd following the partial release of data from the trial. — Preceding unsigned comment added by 146.90.86.66 (talk) 03:36, 27 November 2016 (UTC)
 * The way it works on Wikipedia is explained at WP:WEIGHT. It would be very beneficial for this discussion if those like yourself who don't understand Wikipedia's sourcing policies familiarise themselves with them before attempting to comment further. Where a high quality type of source, like a Cochrane review, reaches a conclusion such as "no evidence suggests that exercise therapy may worsen outcomes", we follow the guidance "Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in the published, reliable sources." That means that the Cochran review's viewpoint on ET worsening outcomes will be reported along with any other equally authoritative sources that have a view on the issue. It does not mean that a bunch of anonymous editors can decide that they don't like the way a review was conducted and can exclude it. That's not critical thinking, that's a recipe for anarchy and POV-pushing. Wikipedia does indeed discourage editors thinking that their analysis of data supersedes that of reliable sources. An expert here is defined as somebody who knows what the best sources are. So, please check the review you adduce: as far as I can see, Marques et al, 2015 looks at the effectiveness of different interventions and says nothing conclusive about harms as you seem to think. If you believe the proposed summaries are misleading, then please explain how we should be summarising a conclusion that states "no evidence suggests that exercise therapy may worsen outcomes". If you are asserting that the conclusions of the Cochrane review are misleading, then name for us the equally reliable source which states that those conclusions are misleading. If you haven't grasped WP:No original research yet, that means we go by what reliable secondary sources tells us, not what amateur analysis by editors might come up with. --RexxS (talk) 04:37, 27 November 2016 (UTC)

Oppose I do not think any of the proposed edits should be included in the lead, although the material being discussed seems be worth covering, at some greater length, in the section on Exercise therapy. Compare this discussion with the article on Major depressive disorder, where the lead briefly mentions treatments, including medication and electro-convulsive therapy, without needing to state that there is 'no evidence of harm', or similar. Wayne 14:49, 5 December 2016 (UTC)
 * Major depressive disorder devotes one out of the four paragraphs of its lead to treatment. The summary of treatment contains six sentences, including "It is unclear whether medications affect the risk of suicide", and six references. Your assertion that "the lead briefly mentions treatments" is simply untrue, and the lead does state where there is no evidence of a possible harm as a result of treatment. Were you thinking of a different article? --RexxS (talk) 18:18, 5 December 2016 (UTC)
 * My 'assertion' that the coverage of treatments in that article's lead is 'brief' is not "simply untrue"; it is in fact how I read the relative emphasis given to the material. As you say, six sentences. I did not interpret the statement on risk of suicide in the same light, but on re-reading it, I concede that interpretation. Nonetheless my 'Oppose' stands. Wayne 23:30, 5 December 2016 (UTC)
 * Of course it still stands. It's just that whoever closes this discussion is now going to see that your opposition is based on your premise of a comparison with "Major depressive disorder "where the lead briefly mentions treatments", according to you. That article devotes six sentences and six references to treatments; this article devotes two sentences and two references to treatments. You oppose expanding the content on treatments in the lead of this article. It expect the closer to give your reasoning the weight it deserves. --RexxS (talk) 00:18, 6 December 2016 (UTC)


 * Oppose for similar reasons as Wayne's and prior editors. Exercise therapy's side effects (or lack thereof in this case) go beyond the scope of the lead. Obviously this should be part of the body text, but remember with giving weight to covering scientific studies that the newest stuff still requires time for counter or supporting studies to come about. SamuelRiv (talk) 19:59, 15 December 2016 (UTC)

Discuss
I remember reading a bit about this some months ago, when the activists were outraged at the recommendation. My (possibly faulty) impression was that it's a bit more complicated than that:
 * Statistically, on average, graduated exercise is good.
 * But that only means that it helps more people than it hurts (or more people than other options). It doesn't mean that it won't hurt anyone.  Exercise for a CFS/ME patient may be the equivalent of chemotherapy for a cancer patient:  Cytotoxic chemotherapy saves lives, but it kills people, too.
 * Exercise is good in the medium- to long-term. Nobody says that CFS/ME people never feel bad during or shortly after exercise.  (In fact, lots of people feel "worse" in some respects the day after exercising, especially if they've been sick and haven't felt up to exercising in recent months.)
 * CFS/ME may not be a single disease, and it's certainly not a single experience, and therefore exercise may help (or hurt) one group of CFS/ME patients, but not another group. It's possible, for example, that graduated exercise is almost always helpful and non-harmful for a person with a mild, recent-onset CFS/ME, but that the same program would be harmful for the rare person with long-standing and severe CFS/ME.

So with these thoughts in mind, I'm slightly concerned about the statement that "it doesn't make people worse". I wouldn't insist that this one sentence explain everything, though, as this is a bit more than the lead should handle. But perhaps a different phrasing would be more precise. WhatamIdoing (talk) 06:23, 26 November 2016 (UTC)
 * Well evidence does not support harm from exercise therapy. This is notable as many claim harm from it. And feeling bad is different than making people worse in the long term. Doc James  (talk · contribs · email) 06:36, 26 November 2016 (UTC)
 * "Does not make people worse in the long term" is not the same as "does not make people worse". Arch (talk) 06:51, 26 November 2016 (UTC)
 * Added a second option. Doc James  (talk · contribs · email) 06:57, 26 November 2016 (UTC)


 * My feeling is that proper explanation and qualification is needed, and all that would suggest a separate, later section, perhaps one dedicated to graded exercise therapy. This is an opening paragraph. I would rather not see treatments discussed at all at this stage. But if they are the statement needs to be very brief: the presence/absence of adverse effects of a treatment requires care and caution. I also do not want to encourage patients to go out and exercise madly, that's reckless. The statement in the Cochrane review is meant to be interpreted within the context of certain specific supervised programmes, and even then they never claim there is "no risk". I am unclear why we are still discussing this, as the overwhelming consensus is that the statement should not appear.--Wilshica (talk) 07:27, 26 November 2016 (UTC)
 * I agree. The Cochrane review on exercise therapy was already included and discussed in its own section of article, before the recent addition by Doc James as an additional reference in the opening paragraph. Arch (talk) 07:56, 26 November 2016 (UTC)
 * Yes the Cochrane review seems to have been used in the section on Exercise therapy, smeared with an unsupported piece of original research: "The evidence for this however comes only from studies of people who are well enough to attend outpatient clinics and that this in turn limited who could be recommended exercise therapy as an intervention." Where is that stated in a reliable source? The section also omitted the first conclusion from the review - I wonder why that is? --RexxS (talk) 23:06, 26 November 2016 (UTC)

Both supporting options are unacceptable. Option 2 is even worse than the original Option 1.

Cherry picking poorly evidenced statements from a systematic review while ignoring the context given in the same source is not consistent with WP:MEDRS. Misinformed content in relatively high profile articles also puts the health of patients at risk. But a limitation of WP:MEDRS is that it does tend to encourage a blind adherence to Cochrane reviews, which is a serious issue when such reviews can merely reflect the biases inherent in the primary studies. Experts such as John Ioannidis are concerned that "evidence-based medicine has been hijacked" and Cochrane reviews "may cause harm by giving credibility to biased studies of vested interests through otherwise respected systematic reviews". Figure 2 of the Cochrane review of exercise for CFS documents the high risk of bias in exercise studies. James Coyne has expressed concerns about conflicts of interest present in the Cochrane review in question.

The Cochrane review in question is specifically referring to serious adverse events (e.g. hospitalisation and death), and the evidence is poor because most trials did not record adverse events in general and when they did it was not reported adequately. The Cochrane review itself states that "sparse data made it impossible for review authors to draw conclusions". In the NIH P2P systematic review is a similar conclusion: "Harms were rarely reported across studies (insufficient evidence)." Older reviews make similar statements. It is negligent to omit this context when placing a questionable and controversial statement at the top of the Wikipedia article: "Exercise therapy does not appear to make people worse." The proposed alternative is worse: "There is no evidence of harm from exercise."

WP:MEDRS states that reports from US government agencies such as the Agency for Healthcare Research and Quality provide trustworthy mainstream views for health content on Wikipedia. This is what a recent AHRQ report (Diagnosis and Treatment of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome) has to say about the current subject : ''"Harms were not well reported, although in one trial patients receiving GET reported more adverse events compared with those receiving cognitive behavior therapy (CBT), adaptive pacing, or usual care; one trial reported more withdrawals of patients receiving GET, one trial had a high percentage of patients refusing repeat exercise testing, and several other trials reported more withdrawals of patients receiving GET, all compared with controls." "Although harms were not well reported across trials, GET was associated with a higher number of reported harms and withdrawals in some trials." "Studies also need to report harms more completely to help identify patients negatively affected by certain treatments."''

(The AHRQ recently published an addendum that downgrades the general quality of evidence for CBT and GET to "low strength of evidence" and "insufficient evidence" for various outcome measures once studies that use the poor quality Oxford criteria are excluded, as such studies may not reflect CFS patients.)

Reports of harms from exercise are very common outside clinical trials, on the order of 50-75% of those surveyed. The number of patients who have reported harms from exercise therapy is much greater than the number of patients who participated in trials that recorded harms. Symptoms becoming worse after exertion is a characteristic feature of any serious definition of ME or ME/CFS, and exercise stress tests show symptoms and biological parameters worsening abnormally after exercise; these facts prompted the Institute of Medicine to suggest a name change to Systemic Exertion Intolerance Disease.

The claim that a medical condition characterized by intolerance to exertion can be safely treated with graded increases in activity is an extraordinary claim that is contradicted by the available evidence: there is no good evidence that CBT or GET safely increases activity in the medium or long term. Evidence from a meta-analysis and a large trial  shows no increases of fitness or objectively measured activity. Therefore wording of the first sentence is also misleading and should be changed: "Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases." It is important to distinguish between the lofty goals of GET and the actual achievements of patients who have done GET. ME/CFS patients are generally unable to sustain increases in activity when instructed to do so; claiming otherwise, against the evidence, is essentially quackery.

_ Tekaphor ( TALK ) 12:53, 26 November 2016 (UTC)


 * The sky is pink. -Roxy the dog. bark 11:53, 26 November 2016 (UTC)
 * The review you link to says "Although CBT effectively reduced fatigue, it did not change the level of physical activity."


 * Not sure why one would think counselling would change fitness levels? It did lead to improvements just not in this aspect. All people have trouble increasing activity levels for a sustained period of time when instructed to do so, nothing special about CFS there. Doc James  (talk · contribs · email) 18:38, 26 November 2016 (UTC)


 * 5BX exercises got Canadian air crews up to super-fit by taking it slowly without the trouble that you speak of. Why keep injecting non sequiturs? You appear to be now flogging a dead horse. So dead, that editors are politely suggesting that the lifeless carcass or cadaver of your argument is now beginning to smell.--Aspro (talk) 19:56, 26 November 2016 (UTC)
 * Yup that exercise program is supported by sources such as "celebsnow.co.uk" :-)  Doc James  (talk · contribs · email) 21:28, 26 November 2016 (UTC)


 * Are you agreeing with me or trying to take a side-wipe by bring up the web site Celebs Now as if you can belittle me by referencing (and there by trying to associate me with popular twaddle) to such a web site. You may have had the opportunity and profession privilege to dig bullets out of people but William's brother served twice in Afghanistan where he was a prime target. To atempt to downgrade and ridicule  5BX, (like you have done above)  is like ........  P.S. in my late twenties my resting heart rate was still 54-56 and lower than a cardiac surgeons heart rate so don't try an' bull-shit me any more. --Aspro (talk) 22:36, 26 November 2016 (UTC)


 * The claim that a medical condition characterized by intolerance to exertion can be safely treated with graded increases in activity is an extraordinary claim that is contradicted by the available evidence. That's simply untrue. You have produced no reliable secondary source that finds that graded exercise therapy is unsafe as a treatment. You've cherry-picked one cell of a table in the AHRQ report which notes that in one trial patients receiving GET reported more adverse events, but the AHRQ report draws no conclusions whatsoever about the safety of GET as a treatment. So what evidence do you have to validly contradict the Cochrane report conclusion, "no evidence suggests that exercise therapy may worsen outcomes"? The answer is "none". --RexxS (talk) 23:25, 26 November 2016 (UTC)


 * I would be greatful if people would actually read the Cochrane review and the context before forming their opinions. It is not the quality of the review that it is being questioned, but the proposed points will be quoting the article out of context. The Cochrane review specifically states "but sparse data made it impossible for review authors to draw conclusions" on the point of adverse effects. This means it is deliberately misleading to include a non-specific point that implies the treatment has no adverse effects. Any point will need to specifically state something along those lines, eg use words along the lines of "sparse data" or "lack of data" or "insufficient data to conclude". Arch (talk) 04:14, 27 November 2016 (UTC)
 * And I would be grateful if you stopped trying to impose your own views on the Cochrane review and let the authors speak for themselves rather than deciding that you know better than they do. The quality of the review is the ONLY thing that you can question per WP:MEDRS. The proposed points will not be quoting the article out of context because those are the authors' own conclusions and the context is the section of the review labelled "AUTHORS' CONCLUSIONS". The Cochrane review states precisely "Serious adverse reactions were rare in both groups (RR 0.99, 95% CI 0.14 to 6.97; one study, 319 participants; moderate-quality evidence), but sparse data made it impossible for review authors to draw conclusions." So the sparse data is referring only to serious adverse reactions, which we all agree are rare - hence 'sparse', but does not say the data on other adverse outcomes is sparse. You are the only one quoting out of context. The authors do not qualify their generalised conclusion, and we should not be doing so either. It is deliberately misleading to take a comment about rare events like "serious adverse reactions" and pretend it applies to a conclusion about all potential adverse effects. --RexxS (talk) 04:55, 27 November 2016 (UTC)
 * Again, please read the article and look at what was reviewed. No other measures of adverse effects were reviewed. The only specific adverse effects reviewed in the article were the "serious adverse reactions" we have been discussing this whole time. Arch (talk) 05:05, 27 November 2016 (UTC)
 * Yet again, please read WP:MEDRS and try to understand why we don't let editors' own views and conclusions dictate content. Anybody can read what the review said about adverse effects in the MAIN RESULTS summary: Serious adverse reactions were rare in both groups (RR 0.67, 95% CI 0.11 to 3.96). We observed little or no difference in physical functioning, depression, anxiety and sleep, and we were not able to draw any conclusions with regard to pain, self-perceived changes in overall health, use of health service resources and drop-out rate. The authors reviewed: (1) physical functioning; (2) depression; (3) anxiety; (4) sleep; (5) pain; (6) self-perceived changes in overall health; (7) use of health service resources; and (8) drop-out rate. They found little or no difference between the ET group and the controls regarding the first four adverse reactions, and were unable to draw conclusions on the latter four. They summarised that in the AUTHORS' CONCLUSIONS section as Patients with CFS may generally benefit and feel less fatigued following exercise therapy, and no evidence suggests that exercise therapy may worsen outcomes. What problem are you having with reading what the authors wrote about the multiple possible adverse reactions? --RexxS (talk) 05:27, 27 November 2016 (UTC)

Added option 3, which is most consistent with what is presented in the Cochrane review. Arch (talk) 04:28, 27 November 2016 (UTC)


 * I've added option 4, which is even more consistent with what is presented in the Cochrane review. I've also respected WP:ASF which tells us to write in Wikipedia's own voice when summarising the view of a reliable source that is not contradicted by an equal source, and to reserve attribution for where different sources hold different views. --RexxS (talk) 05:10, 27 November 2016 (UTC)


 * Arch, it seems you are correct that serious adverse reactions (SARs) are the only specific, individual adverse effects reviewed. Under "Primary outcomes", it lists "Adverse outcomes: measured using any reporting system (e.g. serious adverse reactions (SARs) (European Union Clinical Trials Directive 2001))." Then under each section of the review it has a subsection on "Adverse effects" but only SARs are covered. Then under "Summary of main results" in the "Discussion" section it repeatedly states that it is difficult to draw conclusions about this issue, and this point is repeated in the Abstract. This is consistent with multiple systematic reviews conducted over the past ten years which conclude that harms are poorly reported in general.


 * However, as RexxS seems to be saying, the statement about there being no evidence that exercise therapy worsens outcomes could be based on the other outcome measures reported. If that is what the Cochrane review authors are saying, then we should be clear in the Wikipedia article that this is what they mean. Unfortunately, most of the other outcomes received a quality rating of either very low or low, so their conclusion is still based on low quality evidence according to the authors. Also almost all of the other outcomes are group averages of continuous scores, not binary outcomes for individuals, and such outcomes simply do not detect the number of adverse events for individuals. That said, RexxS's Option 4 is a significant improvement over Option 1 and Option 2, but I think further clarification or caveats is needed. -- Tekaphor  ( TALK ) 09:18, 27 November 2016 (UTC)


 * Doc James, the version of CBT studied in the meta-analysis was not simply counselling but included graded activity/exercise. Also the most studied version of CBT for CFS encourages graded activity/exercise, as avoidance of activity is speculated to be one of the main maladapative behaviours arising from patients' unhelpful beliefs that exercise makes them worse. We should certainly expect such a therapy to increase activity if it works as claimed, as it is part of the process and goal of the therapy. Just because people in general have trouble increasing activity does not mean we should expect such a dedicated programme to completely fail to produce any significant increases in activity, unless of course the target population are unable to do it: CFS patients are generally less active than sedentary controls, they commonly report a low "activity ceiling", intolerance to exertion is a hallmark, and exercise testing shows abnormalities. -- Tekaphor  ( TALK ) 09:18, 27 November 2016 (UTC)


 * RexxS, the available evidence of objective outcomes clearly shows that CFS patients have not increased their fitness or activity during or after graded exercise therapy. This comes from a meta-analysis of three CBT(/GET) trials, two additional large trials of CBT, GET, or pragmatic rehabilitation, and a few smaller trials of activity management. I am not aware of relevant evidence that contradicts this. Perhaps there is a high-quality secondary source summarising all this. If graded increases in activity are not actually occurring, then it is an unsupported leap of faith to claim such increases are safe.


 * For randomised trials of ambulatory CFS patients meeting loose criteria, GET enrollment does not appear to cause a significant increase of SAEs (e.g. hospitalisation and death) or worse outcomes on average for the entire group, but there is no objective evidence of increased exercise, activity, or fitness, while most trials do not properly record or report individual harms so the evidence is limited. The types of harms from exercise therapy reported by patients outside trials are generally not recorded in trials either. The purpose of me quoting the AHRQ report was not to show that GET is unsafe but to further show that harms or adverse effects are poorly reported in the research literature. -- Tekaphor  ( TALK ) 11:25, 27 November 2016 (UTC)


 * I disagree with RexxS's interpretation that "sparse data" means "low number of SAEs". Sparse data is likely referring to the general lack of data on the subject, not the low number of SAEs. Why would a low number of SARs make it difficult to draw conclusions? If anything the low number of SARs would only serve to show it can be concluded that exercise therapy is safe. But the authors state it is "impossible" to draw conclusions, because of sparse data. -- Tekaphor  ( TALK ) 09:18, 27 November 2016 (UTC)
 * Tekaphor, If graded increases in activity are not actually occurring, then it is an unsupported leap of faith to claim such increases are safe. You are making the mistake of confusing efficacy with safety and your conclusion is a non-sequitor. As a simple analogy, nobody but a handful of cranks believes that homeopathy is effective, but nobody believes it is unsafe, either. I don't need a leap of faith to accept that ET doesn't worsen outcomes compared with other treatments, I just need to read the authors' conclusions in the Cochrane review, as you should.
 * Why should sparse data mean anything other than what it is? Serious adverse reactions are a rare occurrence - surely we all agree on that? The implications for trials are clear: you need massive sample sizes to be able to do any meaningful statistical comparisons between two groups when the event you are comparing is rare; the usual sample sizes make it impossible to draw conclusions. That's just stats 101. If two samples each of size 10,000 are made of treatment A and treatment B, and 1 SAR occurs in the A group, while 2 SARs occur in the B group, can you say that treatment B is less safe than treatment A? No, the difference in frequencies is less than 1 standard deviation and that's not significant. Whereas if two samples each of size 10,000 are made of treatment A and treatment B, and 100 SARs occurs in the A group, while 200 SARs occur in the B group, can you say that treatment B is less safe than treatment A? Of course you can - the difference between the frequencies is over 8 standard deviations. Now do you understand why a low number of events ("sparse data") make it difficult to draw conclusions on any comparison?
 * But I'm not trying to read anything into the sparse data of SARs, unlike your colleague. I am merely concerned by the rejection of the conclusions concerning all adverse outcomes of one of the highest quality type of sources we have by editors who do not bring forward sources that disagree. These editors merely argue ad tedium that the authors of the Cochrane review didn't meant what they wrote: "no evidence suggests that exercise therapy may worsen outcomes." Please re-read what I quoted above about WP:WEIGHT. If Wikipedia policies and guidelines continue to be flouted like that, at some point soon this will tip over from a content dispute into a behavioural one. --RexxS (talk) 21:45, 27 November 2016 (UTC)
 * But I'm not trying to read anything into the sparse data of SARs, unlike your colleague. I am merely concerned by the rejection of the conclusions concerning all adverse outcomes of one of the highest quality type of sources we have by editors who do not bring forward sources that disagree. These editors merely argue ad tedium that the authors of the Cochrane review didn't meant what they wrote: "no evidence suggests that exercise therapy may worsen outcomes." Please re-read what I quoted above about WP:WEIGHT. If Wikipedia policies and guidelines continue to be flouted like that, at some point soon this will tip over from a content dispute into a behavioural one. --RexxS (talk) 21:45, 27 November 2016 (UTC)

I'm not confusing efficacy with safety. Don't confuse a planned programme with actual achievements. Data from trials are collected from patients who enrolled and had sessions with the therapist, but the objective evidence of activity and fitness from these trials shows a failure to facilitate increases of activity and fitness. That's why the available evidence doesn't support the blanket claim that increases of activity per se are safe. None of the cited studies in the top section of the Wikipedia CFS article provide evidence of increased activity; they only describe that increases of activity are planned. Hopefully the Cochrane review authors will follow through on their positive response to the person who requested that objective outcomes should be analyzed too.

Based on limited evidence ranging from very low to moderate quality, ambulatory patients meeting loose CFS criteria who participate in a GET programme do not have worse outcomes on average and serious adverse reactions are rare, but the available data are limited. You mention almost no one believes homeopathy is harmful; that may be so, but it also has no credible mechanism for harm (nocebo and toxic impurities aside), whereas ME-CFS has credible mechanisms of harm and good reason to doubt blanket claims about safely increasing activity.

It's true that large sample sizes are required to detect significant differences between groups when the event frequency is rare, good point, but it's not really clear that's why the authors are saying "sparse data made it impossible for review authors to draw conclusions", especially given the context that almost all reviewed trials didn't record or report these outcomes. Anyway, moving on to the main issue...

To all, we need a consensus on what the authors mean by exercise therapy does not worsen outcomes, how to best word it for Wikipedia, and include appropriate caveats about the quality of the evidence. I am open to the interpretation that they generally mean all outcomes on average instead of SARs, but it's not entirely clear this is the case so disagreement was unsurprising. To those who previously opposed the suggestions, what are your thoughts now? Do you agree with RexxS that the authors' statement about exercise not worsening outcomes is referring to outcomes in general? Or is it referring to SAEs, or something else like fatigue? What caveats about the evidence should be included in the Wikipedia article? The relevant sections are under "Authors' conclusions" and "What does evidence from the review tell us?" . I see Arch now supports Option 4. -- Tekaphor  ( TALK ) 05:09, 30 November 2016 (UTC)
 * I'm afraid you are confusing efficacy with safety: the objective evidence of activity and fitness from these trials shows a failure to facilitate increases of activity and fitness is exactly the mistake you're making. Any failure to facilitate increases of activity and fitness is a failure of effectiveness, not of safety. It is clear to most that failure to achieve an outcome is not the same as a failure to be safe. Claims made by the best reliable sources are "claims", pure and simple, not "blanket claims" or anything else. In Wikipedia, they are "facts", unless contradicted by an equally reliable source. As it happens, there may be a credible mechanism for harm in using homeopathic treatments (getting the dosage wrong), but that's not the issue. MEDRS sources are needed to say anything about mechanisms and you don't have any to support your assertion that ME-CFS has credible mechanisms of harm and good reason to doubt blanket claims about safely increasing activity, do you?
 * Again you think that you can drill down into a Cochrane review and second-guess what the authors read and their analysis. You can't. A Cochrane review is not "limited evidence ranging from very low to moderate quality". It's the highest quality type of source we have and it does state quite conclusively that those who participate in ET programmes do not have worse outcomes (compared with other forms of intervention). I make that qualification because in this discussion it's not always seemed obvious that the Cochrane authors were looking for differences between those undergoing ET and those receiving other treatment. There was not a "no intervention" option in the comparison for obvious reasons. So if you're asking "is ET as safe as doing nothing?", the answer is nobody knows - it's untested in the secondary sources. But it's abundantly clear that the Cochrane authors found there was sufficient evidence available on four areas of non-severe adverse reactions (physical functioning, depression, anxiety and sleep) to draw the conclusion that ET did not worsen them compared with the other interventions reviewed. The "sparse data" caveat that others have made so much of is unquestionably applied only to the conclusions concerning severe adverse reactions: "Serious adverse reactions were rare in both groups ... but sparse data made it impossible for review authors to draw conclusions." is the only place the phrase occurs.
 * You could quite possibly make a case that the conclusions should be qualified by naming the four areas where no difference in worsening were found, but I'd advise against it. The Cochrane authors did not feel it necessary to so do in their summary conclusions. Furthermore, to be blunt, I think that doing so in this article would be putting a spin on the text simply because some editors find it difficult to believe that there is no high quality evidence of worsened outcomes, and want to hedge what is stated here to conform with their world-view. I'm sorry, but sources and sources alone determine content on Wikpedia, much as we'd wish it different on occasion. --RexxS (talk) 15:50, 30 November 2016 (UTC)
 * Again you think that you can drill down into a Cochrane review and second-guess what the authors read and their analysis. You can't. A Cochrane review is not "limited evidence ranging from very low to moderate quality". It's the highest quality type of source we have and it does state quite conclusively that those who participate in ET programmes do not have worse outcomes (compared with other forms of intervention). I make that qualification because in this discussion it's not always seemed obvious that the Cochrane authors were looking for differences between those undergoing ET and those receiving other treatment. There was not a "no intervention" option in the comparison for obvious reasons. So if you're asking "is ET as safe as doing nothing?", the answer is nobody knows - it's untested in the secondary sources. But it's abundantly clear that the Cochrane authors found there was sufficient evidence available on four areas of non-severe adverse reactions (physical functioning, depression, anxiety and sleep) to draw the conclusion that ET did not worsen them compared with the other interventions reviewed. The "sparse data" caveat that others have made so much of is unquestionably applied only to the conclusions concerning severe adverse reactions: "Serious adverse reactions were rare in both groups ... but sparse data made it impossible for review authors to draw conclusions." is the only place the phrase occurs.
 * You could quite possibly make a case that the conclusions should be qualified by naming the four areas where no difference in worsening were found, but I'd advise against it. The Cochrane authors did not feel it necessary to so do in their summary conclusions. Furthermore, to be blunt, I think that doing so in this article would be putting a spin on the text simply because some editors find it difficult to believe that there is no high quality evidence of worsened outcomes, and want to hedge what is stated here to conform with their world-view. I'm sorry, but sources and sources alone determine content on Wikpedia, much as we'd wish it different on occasion. --RexxS (talk) 15:50, 30 November 2016 (UTC)
 * You could quite possibly make a case that the conclusions should be qualified by naming the four areas where no difference in worsening were found, but I'd advise against it. The Cochrane authors did not feel it necessary to so do in their summary conclusions. Furthermore, to be blunt, I think that doing so in this article would be putting a spin on the text simply because some editors find it difficult to believe that there is no high quality evidence of worsened outcomes, and want to hedge what is stated here to conform with their world-view. I'm sorry, but sources and sources alone determine content on Wikpedia, much as we'd wish it different on occasion. --RexxS (talk) 15:50, 30 November 2016 (UTC)


 * But I'm not arguing that the failure to increase activity is a failure of safety, I've been saying all along that the failure to increase activity means it can't be concluded that increases of activity per se are safe. Going back to the chemotherapy comparison someone used earlier, imagine if patients were enrolled in a programme that encouraged graded increases of chemotherapy, but objective evidence (of efficacy) shows that the actual amount of chemotherapy administered basically remains unchanged; we can't then claim that increases of chemotherapy are safe, we can only make evidence-based conclusions about the participation in the chemotherapy encouragement programme.


 * This argument is secondary to the RfC and I only raised it because the introduction of the Wikipedia article (possibly other instances) says "a gradual increase in activity suited to individual capacity can be beneficial in some cases." That said, the part about suited to individual capacity is an important caveat. But IIRC the cited sources don't actually say activity increases occurred, they only describe the activity increases as planned, generally once a safe baseline is established. My point is that we must be careful to stick to the wording in the cited sources; suggesting increases of activity actually occurred is likely to be WP:OR unless the authors of the sources in question (incorrectly) say activity increases occurred.


 * By "credible mechanism of harm" I didn't mean a high-quality secondary source overriding the Cochrane review. The credible mechanism that exercise therapy can (potentially) harm is that over-exertion is commonly known to exacerbate the symptoms (this is even mandatory in some case definitions) and exercise tests worsen physiological parameters beyond what happens to sedentary controls. In trials of exercise programmes for CFS, special care is taken to avoid doing too much too quickly, and this care wouldn't be necessary if there was no potential for harm. A "blanket claim" is a generalization, an appropriate term when making a generalized claim about safety without caveats. -- Tekaphor  ( TALK ) 10:50, 1 December 2016 (UTC)

I noticed the RFC today. Since its still active I'll give my 2 cents. All 4 options listed above sound awkward for the lead. It currently reads "Evidence suggests that cognitive behavioral therapy and a gradual increase in activity suited to individual capacity can be beneficial in some cases." I think that sums it up nicely. Cochrane is excellent for providing quality systematic reviews, but we don't want the lead to sound like a lawyer drafted it. In my opinion, put the reference as a footnote after "cases" and put any comments in the body.Dig Deeper (talk) 03:28, 8 December 2016 (UTC) Dig Deeper (talk) 03:28, 8 December 2016 (UTC)
 * Just got around to it myself. But just to have some fun I measured it: 8284 characters. A lot of stubs could have expanded to starts with all this energy. Best Regards,
 * Barbara (WVS) (talk)

Reorganization of pathophysiology section
That section is just a total clusterfuck right now. I am going to try to consolidate findings and reorganize it into a more readable unified section.Petergstrom (talk) 01:55, 9 January 2017 (UTC)

Small primary source
Details from this small primary source have been added to the lead a couple of times. I have removed them per WP:MEDRS. We should be using high quality secondary sources. Doc James (talk · contribs · email) 23:24, 15 January 2017 (UTC)

The largest trial into CBT and GET actually showed that these treatments don't improve chances of recovery
There is now a peer reviewed paper on this. [http://www.tandfonline.com/doi/abs/10.1080/21641846.2017.1259724 Can patients with chronic fatigue syndrome really recover after graded exercise or cognitive behavioural therapy? A critical commentary and preliminary re-analysis of the PACE trial] Abmayer (talk) 11:31, 14 February 2017 (UTC)
 * We use reviews generally rather than single trials. Doc James  (talk · contribs · email) 03:09, 15 February 2017 (UTC)

Differential diagnoses subsection
This subsection contains a list formatted as article text, running to a length of around 10 lines. Would someone like to format it as a list, with colums? Note that it is grouped according to system, as well.--Quisqualis (talk) 00:22, 16 February 2017 (UTC)

This topic urgently needs updating
This topic urgently needs updating, and I do not now have suitable references to do it myself.

Internationally and in the U.S., this condition is now most commonly referred to in peer-reviewed medical journals as ME/CFS (or CFS/ME) -- ME for Myalgic Encephalomyelitis, a name widely used for the condition (and recognized by the UN World Health Organization) before the CDC designated it Chronic Fatigue Syndrome. The name Systemic Exertion Intolerance Disease was proposed in a recent National Institute of Medicine magisterial study, but has not been widely adopted.

The assertion that "that cognitive behavioral therapy and a gradual increase in activity ... can be beneficial" is, to be polite, highly controversial. The PACE study that affirmed this conclusion has been heavily criticized by internationally known researchers, but has never been withdrawn. Critics included the International Association for Chronic Fatigue Syndrome/Myalgic Encephalomyelitis [www.iacfsme.org/].

Other resources of note: Trans-NIH Myalgic Encephalomyelitis/Chronic Fatigue Syndrome Working Group Solve ME/CFS  This organization has frequent webinars featuring top researchers that are available on YouTube. Not a good reference for an article, but a quick way to ID researchers. One posted November 2016 featuring Anthony L. Komaroff, MD, Simcox-Clifford-Higby professor of medicine at Harvard Medical School amounts to a review of current knowledge about ME/CFS

Rev.trw379 (talk) 21:22, 11 March 2017 (UTC)


 * There's 19 pages of archived discussions you can access from the box at the top of this page. All your points have been discussed in extenso and I would suggest you check with the archives first before reopening the debate. JFW &#124; T@lk  20:39, 13 March 2017 (UTC)

New CDC Guidance: Post-Exertional Malaise (PEM) is now a Primary Symptom required for diagnosis
Chronic_fatigue_syndrome needs amending to reflect changes in the CDC guidance on symptoms and in particular the 3 required primary symptoms for diagnosis. See https://www.cdc.gov/me-cfs/symptoms-diagnosis/symptoms.html

C7762 (talk) 18:31, 14 July 2017 (UTC)


 * Agree and more so: Think that Post-exertional malaise now needs its own article. One of the problems that Wessely et. al. created by lumping ME together with CFS (unilaterally doctoring the UK's database of ICD 9 without consent), (which apparently enabled him to funnel the whole of the £2 million UK grant money for research through his own departments) is that many  do not suffer  'chronic' fatigue but  exertional exhaustion. Chickens are now coming home to roost and the CFS article needs to be split up again and put back (as it originally was) to the pre  Wessely et. al. era. Currently, it is like having an article that lumps together smallpox, chickenpox and measles because they all produce skin blisters. Aspro (talk) 17:04, 17 July 2017 (UTC)
 * Will need to get caps correct. Doc James  (talk · contribs · email) 04:17, 18 July 2017 (UTC)


 * Think Post-exertional malaise is more inline with our naming policy than 'Post-Exertional Malaise' regarding Caps. The hyphen is necessary though, for grammatical correctness. PEM doesn’t need to be included in the title, since  it is   tautologously redundant and can be handled with an addition to the disambiguation  PME page. Aspro (talk) 17:15, 18 July 2017 (UTC)

GETSET trial
10.1016/S0140-6736(16)32589-2 - this is a primary source but does seem to indicate the safety of self-administered GET. Let the debate begin... JFW &#124; T@lk  11:02, 21 July 2017 (UTC)
 * Seems inline with the 2016 Cochrane review Doc James  (talk · contribs · email) 23:38, 21 July 2017 (UTC)
 * The GETET trial has been heavily criticised for it methodical flaws (for instance no control group, no objective measures and mid trial changes to protocol) - But as Primary sources are not admissible to Wikipedia it is not worth debating. C7762 (talk) 16:12, 28 July 2017 (UTC)
 * Yup we will need to wait to see what the high quality secondary sources make of it. Doc James  (talk · contribs · email) 19:13, 28 July 2017 (UTC)

The CDC guide has been updated.
The CDC guide has recently been updated and no longer recommends Exercise Therapy or CBT as treatments for CFS. I have modified the Management section to reflect this.

https://www.cdc.gov/me-cfs/treatment/index.html

C7762 (talk) 09:28, 8 July 2017 (UTC)

The CDC guide now states "There is no cure or approved treatment for myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)." I will look at incorporating this into the management section.

C7762 (talk) 09:33, 8 July 2017 (UTC)
 * Sounds good. Doc James  (talk · contribs · email) 04:31, 9 July 2017 (UTC)

I propose the following :-

"The CDC guide states "There is no cure or approved treatment". However it does cover methods that might provide some relief such as symptom treatment, diet, activity management, counselling, ongoing care, and disability caused by CFS. "

replaces:-

"The CDC guide covers symptom treatment, diet, activity management, counselling, ongoing care, and disability caused by CFS. "

C7762 (talk) 10:11, 11 July 2017 (UTC)

The CDC guide now opens: 'Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a serious, long-term illness that affects many body systems. People with ME/CFS are often not able to do their usual activities. At times, ME/CFS may confine them to bed. People with ME/CFS have severe fatigue and sleep problems. ME/CFS may get worse after people with the illness try to do as much as they want or need to do. This symptom is known as post-exertional malaise (PEM). Other symptoms can include problems with thinking and concentrating, pain, and dizziness.' I.e., there is less focus on 'fatigue' as the core symptom and more on PEM and sleep in comparison. This is one of the core points of contention for people who don't like the term CFS, so it's important that this is mentioned. Most definitions also mention the 'many body systems' thing, too, so it's relevant to put that in there. Also, both CDC and NICE/NHS describe ME/CFS as a neurological illness, the UK's position confirmed again by the Countess of Mar in the UK House of Lords recently, so that definitely needs to go in there too.

Meanwhile, over in the UK, the MEA and Action for ME (as part of Forward-ME) are pushing for an update like that of the CDC (i.e., removal of CBT and GET as primary treatments), so either that needs to stay in the opening section with the mentions of CBT or GET, or the whole thing needs to be moved to 'Management' (where, TBH, there is already a very good outline of the main issues). It's misleading to put CBT and GET upfront without the balancing info re: the reanalyses of PACE and the removal of CBT/GET from the CDC guide. Wordswordswordsmith (talk) 20:53, 11 July 2017 (UTC)
 * Many of these changes need discussion
 * meassociation.org.uk is not the best source.
 * this is also not a suitable source www.actionforme.org.uk
 * Would need consensus to change it all to ME/CFS
 * Per WP:LEAD the lead should be kept at 4 paragraphs not 7
 * Some of the use of caps were not appropriate
 * We had a recent RFC about GET.
 * Doc James (talk · contribs · email) 00:25, 12 July 2017 (UTC)

I propose a small amendment to wiki/Chronic_fatigue_syndrome to remove any ambiguity on the CDC guidance for CBT and GET.

'From'
 * The CDC states that while there is no cure a number of methods might improve symptoms. The NICE guideline is directed toward clinicians, and specifies the need for shared decision-making between the patient and healthcare professionals, with acknowledgement of the reality and impact of the condition and the symptoms. The NICE guideline covers illness management aspects of diet, sleep and sleep disorders, rest, relaxation, and pacing. Referral to specialist care for cognitive behavioural therapy, graded exercise therapy and activity management programmes are recommended to be offered as a choice to patients with mild or moderate CFS.

'To'
 * The CDC states that while there is no cure a number of methods might improve symptoms. The NICE guideline is directed toward clinicians, and specifies the need for shared decision-making between the patient and healthcare professionals, with acknowledgement of the reality and impact of the condition and the symptoms. The NICE guideline covers illness management aspects of diet, sleep and sleep disorders, rest, relaxation, and pacing. Referral to specialist care for cognitive behavioural therapy, graded exercise therapy and activity management programmes are recommended to be offered as a choice to patients with mild or moderate CFS by NICE whereas the CDC stopped recommending CBT and GET in May 2017.

C7762 (talk) 18:17, 14 July 2017 (UTC)
 * Were does it say the CDC no long recommends CBT or GET?
 * It says "Other pain management methods include stretching and movement therapies, gentle massage, heat, toning exercises, and water therapy for healing... Talking with a therapist to help find strategies to cope with the illness and its impact on daily life and relationships...Avoiding ‘push-and-crash’ cycles through carefully managing activity." Doc James  (talk · contribs · email) 01:32, 15 July 2017 (UTC)

Hi, please consider the following points. In particular I think consensus needs to be reached on point 2.

1. Previous guidance from CDC explicitly recommended CBT and GET. That explicit guidance has been removed. Instead of saying "removed" it would be equally correct to say the CDC does not recommend CBT or GET.

2. I think a lot of the debate about CBT for CFS on Wikipedia is due to crossed wires. CBT as delivered to CFS patients, in the NHS for example, (is "by mental health professional, who has ... training in applying CBT to patients with CFS/ME") is not CBT as most people would understand it to be. And there is little doubt normal CBT would be of great benefit in helping many CFS patients deal with their illness. But CFS-CBT is not about coping statagies, it works on the underlying Theoretical Model (from  the PACE trial CBT manual*):-
 * " Fear of symptoms and consequent avoidance of activity associated with symptoms is central. ... The model assumes that physiological (fatigue), cognitive (fear of engaging in activity) and behavioural responses (avoidance of activity), are linked. Therefore by modifying one response it is anticipated that changes occur in the other responses. For example, increasing activity (behaviour) may gradually reduce the fear (cognitions) that activity leads to worsening of symptoms.".

It assumes CFS is maintained by illness beliefs not any underlying biological mechanism. Unless consensus can be reached that CFS-CBT is different from CBT then I fear debates on the subject on Wikipedia will keep going round in circles.

3. "Talking with a therapist to help find strategies to cope with the illness and its impact on daily life and relationships" describes various methods of counselling or talking therapies (and which could describe supportive CBT helping find coping strategies) but does not describe CFS-CBT as described in the PACE CBT manual or as implemented. It is like saying "various medicines will help diabetes" equates to "Tramadol recommended for diabetes".

4. "stretching and movement therapies,toning exercises" is not Graded Exercise therapy as recommended previously by the CDC or currently by NICE and implemented by say the NHS. GET is tightly defined in the PACE GET manual. The underlying Theoretical Model :-
 * "The model emphasises the importance of the participant’s understanding of their illness and their interpretation of symptoms. For example, they may interpret symptoms as a warning sign to reduce activity. Fear of symptoms and consequent avoidance of activity associated with symptoms is central. ... The model assumes that physiological (fatigue), cognitive (fear of engaging in activity) and behavioural responses (avoidance of activity), are linked. Therefore by modifying one response it is anticipated that changes occur in the other responses. For example, increasing activity (behaviour) may gradually reduce the fear (cognitions) that activity leads to worsening of symptoms. "

5. "Avoiding ‘push-and-crash’ cycles through carefully managing activity" better describes the various forms of Pacing than it does GET.

6. There is a direct link from the PACE manuals to the recommendations.The Cochraine review which fed into the NICE recommendations relied most heavily on the PACE trial (Being by way the largest trial of CBT and GET in CFS). The PACE trial manual for CBT Threapists :- http://www.wolfson.qmul.ac.uk/images/pdfs/3.cbt-therapist-manual.pdf and the GET manual :- http://www.wolfson.qmul.ac.uk/images/pdfs/5.get-therapist-manual.pdf. If it was to be argued that the manuals do not reflect the guidance then it should be presumed the findings of the PACE trial were not feed into the guidance decision making process.

C7762 (talk) 10:59, 15 July 2017 (UTC)
 * The CDC source does not say they do not recommend CBT or GET. They simply do not mention GET by those exact words. Doc James  (talk · contribs · email) 11:30, 15 July 2017 (UTC)


 * To Clarify I am not saying they are opposing GET but they are definitely not recommending it other wise they would state so. But they have removed it from their guidance as previous CDC guidance did explicitly recommend it - You could say they have dropped it. They do not mention it by name or any description which is understood to be GET. I am referencing the therapist manuals to obtain the definition of GET as used in the studies which the guidance is based on. Are you saying GET as recommended by NICE is not was studied by the PACE Trial?


 * To make sure we are not crossing wires, please briefly describe what you understand GET to be in relation to CDC & NICE guidance and in the treatment of CFS. My understanding is that it is based on the Theoretical Model previously described and involves lowering activity (and 'breaking the doom and bust cycle') until symptoms have stabilised then incrementally and gradually increasing exercise (commonly walking) regardless of symptoms, as symptoms are the result of fear of activity. That GET will eventually result in an improvement or cure as the fear of activity has been removed. It should not be confused with general "stretching and movement therapies,toning exercises" or general activity management or Pacing. C7762 (talk) 13:53, 15 July 2017 (UTC)
 * We do not mention GET in the lead either. Doc James  (talk · contribs · email) 00:41, 16 July 2017 (UTC)
 * It should also be noted that the opening paragraph to the CDC guidance Treatment section states: "There is no cure or approved treatment for myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). However, some symptoms can be treated or managed. The CDC does not recommend any treatment for the underlying CFS but that relief of some symptoms may be possible (such as pain or sleep). C7762 (talk) 14:52, 15 July 2017 (UTC)
 * When they refer to approval they only mean FDA approval. Not a global thing. But I agree we should say no "cure" and added it already a number of days ago Doc James  (talk · contribs · email) 00:41, 16 July 2017 (UTC)
 * The CDC site still says "Some people with CFS find graded exercise or activities helpful." and then continues to give more specific details, so I'm sure why you think they've dropped it, or are not recommending it. --sciencewatcher (talk) 16:35, 15 July 2017 (UTC)


 * Please can you put a reference to where it says that as I have just searched through https://www.cdc.gov/me-cfs/index.html (and the pages under it )which is the most up to date guidance (which was updated 03/07/17) and I can not find the phrase  "Some people with CFS find graded exercise or activities helpful.". or as I have stated before any reference to "Graded Exercise" C7762 (talk) 17:42, 15 July 2017 (UTC)


 * You appear to be referencing out of date guidance https://www.cdc.gov/features/chronic-fatigue-syndrome/index.html which was last updated on 30th July 2014. At the bottom of the page under More Information provides a link ("CDC's Chronic Fatigue Syndrome website" )to the more up to date (3 July 2017) guidance. Searching for "Chronic Fatigue Syndrome" on both the CDC website and in google only brings up the latest guidance. And I only was able to navigate to it by going back to the archive of 2014 pages on the CDC website. C7762 (talk) 19:58, 15 July 2017 (UTC)


 * Actually, both the old and new pages come up with you search for "chronic fatigue syndrome". (The search on the CDC site is just Bing). There's nothing in the old features page that says it's out of date, but it doesn't seem to be linked to from anywhere on the site. I guess they're in the middle of updating it...they still haven't done the "information for healthcare providers" section yet. --sciencewatcher (talk) 21:25, 15 July 2017 (UTC)
 * Will likely take some time before pages like are updated. Appears the CDC is still recommended GET in some places though.  Doc James  (talk · contribs · email) 00:44, 16 July 2017 (UTC)
 * Once the CDC website updates are finished later in the year we should be able to make any relevant changes C7762 (talk) 15:24, 17 July 2017 (UTC)
 * Yup. We will just need to be careful to balance the CDC with other sources and not to go beyond what they state. Doc James  (talk · contribs · email) 04:16, 18 July 2017 (UTC)
 * Said :-
 * "When they refer to approval they only mean FDA approval. Not a global thing."
 * This is an assumption not supported by the text - there is no mention of FDA (which is responsible for drugs) nor does the text say "there is no approved drug". The text clearly states that in the CDC's view there is no approved treatment - it is within the same sentence and tied to "no cure". And this should be reflected otherwise the entry does not reflect the source and could be viewed as having bias. C7762 (talk) 16:24, 2 August 2017 (UTC)
 * When US sources say no approval they nearly always only refer to the US. Let me look at the EU and UK and Canada. Wasn't that one med approved in the EU? Doc James  (talk · contribs · email) 21:20, 2 August 2017 (UTC)
 * For example this is clearer "Fourth, there are no drugs or therapies approved by the U.S. Food and Drug Administration (FDA) to treat ME/CFS."
 * Appears "rintatolimod" has been approved specifically for ME/CFS in Argintina as of 2016 and is the first approval anywhere Doc James  (talk · contribs · email) 22:01, 2 August 2017 (UTC)
 * I may have mis-understood what you originally said - I agree with "there are no drugs or therapies approved by the U.S." - and think this should be included. C7762 (talk) 00:25, 3 August 2017 (UTC)
 * Okay added. Doc James  (talk · contribs · email) 05:17, 3 August 2017 (UTC)