Talk:Chronic fatigue syndrome/Archive 5

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents 1 Overtraining and CFS 2 Alternative medicine 2.1 Alternative medicine and Anti-oxidants 3 This article is about people 4 Unexplained and persistent fatigue 5 Placebo, what placebo? 6 Wilson's temperature syndrome 7 Antidepressants 8 Chinese 9 Chinese 10 Protection 11 More on acupuncture 12 Stress and infection in CFS revisited 13 Thank you, Jagra 14 Other medical treatments section 15 POV changes 16 proposal 17 verify source and request quote 18 alternative medicine 2 19 Attribution of source 20 Wallman et al 21 GET unreferenced statment removed 22 Reliable sources for scientific statements

Overtraining and CFS

Sciencewatcher, Overtraining unlike CFS is associated with high cortisol and therefore not directly comparable to CFS. PMID 12696983 There are some similarities in the cytokine profiles but that does not equate with similar induction, unless you can explain the differences? PMID 14738459 It seems chronic inflammation has been associated with long term brain damage, but more likely to occur with Primary depression because high cortisol in that condition (unlike CFS) restricts regeneration of neuronal networks. PMID 17705097 If there is brain damage / injuries then supplementation with n-3 fatty acids is advisable. PMID 15129302 and have shown benefit in such injury PMID 12640327 This paper links ACTH vasodilation of the adrenal glands with n-3 fatty acid (EPA) release and increased cortisol. PMID 17446179 n-3 fatty acids (EPA) are vasodilative in the brain and reduce inflammatory PGE2 PMID 2141767 PMID 1590442 Fish oil supplementation reduces exercise induced prostaglandin PGE2 and increased cortisol PMID 17923401 and improves blood flow in other exercise intolerant conditions PMID 16770472

Thus n-3 fatty acid has potential benefit in CFS, indeed n-3 fatty acids are found deficient in CFS (likely due to hyper-response) Supplementation with (EPA) has been demonstrated by MRI to improve brain structure in CFS and reduce symptoms PMID 15117099 PMID 16935966 I propose these recent findings be included in the article.Jagra 01:08, 11 November 2007 (UTC)

No, overtraining is associated with high cortisol during the initial stages, follewed by low cortisol during the chronic phase, which is likely similar to CFS. See for example http://www.rrca.org/resources/articles/sum99ots.htm. The CFS fish oil study was done on one patient with no controls, so we don't know if it was actually the fish oil that helped. The MRI doesn't prove that the fish oil caused the changes in the brain - the placebo effect could have caused the change (or else just natural recovery). Basant Puri has been proposing fish oil for years as a treatment for CFS, but so far he has not come up with any controlled studies that actually show it has a benefit, as far as I am aware. And there has been speculation for even longer that fish oils might help schizophrenia and bipolar disorder, but again the evidence isn't very strong. --Sciencewatcher 19:03, 12 November 2007 (UTC)

Interesting but not WP:RS unlike references above, note also different sympathetic function to CFS, so not the same! Do you have any evidence of high cortisol in CFS, at any stage, I'm not aware of it? Fish oil trials, lack of sufficient evidence, is not evidence of lack? prelimunary trials include PMID 15041033, lack of commercial sponsor for a natural product just as likely a reason. A search of pubmed will reveal 139 papers by Puri B, his expertise appears to be in MRI brain studies in conditions with known fatty acid changes, the reason he advocates fish oil is because he has observed and reported many such improvements including ( only recently) with new techniques/tecnology in CFS PMID 16777669 PMID 16777668. I would not shoot the messenger just yet! Have to disagree, there is nothing natural about recovery in CFS. Jagra 02:04, 13 November 2007 (UTC)

Would agree I've never heard of high cortisol in CFS, which isn't say it may be an acute phase phenomenon. But finding similarities between CFS and in this case, overtraining simply on the basis of cortisol and cytokines (whch are quite non specific) is superficial as there are far too many other findings to take into account. Anyway most sources agree that CFS is likely discrete even if it's heteregenous, so attempts to redefine it as effectively misdiagnosed overtraining etc are flimsy (even if, to complicate matters, CFS is as Hyde says, often a misdiagnosis). MEspringal 06:57, 14 November 2007 (UTC)

As stress is a common trigger for CFS (and stress is associated with high cortisol), you could say that CFS patients have high cortisol before they develop CFS. At the moment the research into the HPA axis in CFS patients is confusing, probably because it changes a lot depending on what stage of the illness the patient is in. Overtraining and CFS are similar in that both are triggered by stress. You could say that overtraining is being misdiagnosed as CFS, but then you are making an assumption that CFS is not a stress-related illness. --Sciencewatcher 23:49, 15 November 2007 (UTC)

Sciencewatcher, Even where stress is involved in CFS induction, the evidence indicates it is just as likely due to allostatic load PMID 16610956 PMID 16610958 CFS is considered a Type 4 load which suggests it is a protective adaptive response of the body, become chronic. Where low HPA responsiveness may result in increased activity of other systems including the immune responses and have health implications. See FIG.1 and discussion following in the full text ot this one PMID 10649824. Which means it is already hypocortisol maybe even chronic in some before CFS induction not just starting afterwards as you claim above.Jagra 08:00, 16 November 2007 (UTC)

I'm only aware of one study where CFS patients were more likely than healthy controls to have exercised regularly before illness onset (67% vs 40%) [PMID 8644768]. However, terms like "over-training" and "over-activity" are relative and dependent on individual capacity or circumstance. No one here is proposing that CFS is the same as overtraining syndrome, but even the similarities between them have been overstated; as far as I know, there is no solid evidence that high cortisol is actually involved in either condition's acute phase. Furthermore, overtraining is not merely "triggered by stress" as implied above but is specifically caused by ongoing physical over-exertion, something which is exclusionary for a CFS diagnosis. Of course, someone who assumes that all stress has the same biological effect could argue that chronic psychological stress could be a form of over-activity. Perhaps the studies about the HPA axis in CFS are confusing because the HPA axis simply isn't as fundamentally involved as assumed. There is so much more to exertion than the blood concentration of cortisol. I'm sure we can all concede that over-activity at least could be one of many co-factors for some people's CFS. I agree with Jagra that over-exertion seems to become much more of a problem once CFS has developed, and with Guido that over-exertion prior to CFS could increase susceptibility by lowering defences (or whatever). As for essential fatty acid trials, don't forget about [PMID 2270749]. - Tekaphor 08:08, 16 November 2007 (UTC)

Yes, it is either physical or mental over-exertion that seems to trigger CFS. In the case of overtraining it is physical, and in the case of CFS it is generally mental/emotional, and sometimes illness as well (although it is difficult to tell if the illness is an actual trigger or just another symptom). The common factor in both is excessive stress, which acts on the body in the same way whether it is physical, mental or emotional, or due to illness. In both overtraining and CFS, the symptoms and physiological abnormalities are consistent with stress/burnout.---- Sciencewatcher (talk) 18:19, 16 November 2007 (UTC)

Which specific "physiological consistencies" are you referring to? - Tekaphor (talk) 02:21, 21 November 2007 (UTC)

Abnormalities in the HPA axis and autonomic nervous system...typically a reduced HPA axis drive. --Sciencewatcher (talk) 21:47, 21 November 2007 (UTC)

Sciencewatcher, This study found HPA-axis in burnout normal, hyper-cortisol in primary depression and hypo-cortisol in CFS PMID 16150550 Burnout is characterized by low serotonin or low dopamine, PMID 17641533 Some studies indeed show high cortisol in burnout as might be expected. PMID 12782748 Burnout then would seem to have more in common with Overtraining than CFS. CFS has high serotonin, low cortisol, and there are no studies linking CFS onset with high cortisol. Cortisol levels in infectious mononucleosis have not been found predictive of CFS, PMID 12877399 but there is some evidence that high cortisol levels can cause reactivation of latent Epstein Barr virus PMID 12191645 PMID 11719627 In which case hypo-cortisol might even be seen as a protective strategy! So whilst there may be some overlap in symptoms with other conditions of hypo-cortisol more directly linked to stress, it is still a long bow to draw to conclude that CFS is therefore the same, and "consistent with stress/burnout ". Jagra (talk) 01:12, 25 November 2007 (UTC)

Stress generally causes high cortisol, but after burnout it changes to low cortisol. However as you point out, not all studies show low cortisol in burnout - some show normal cortisol. And the same is true for stress and overtraining. So it is not possible to correlate cortisol with stress or burnout. However what we can say is that abnormal (high or low) cortisol is a pointer that indicates an abnormality in the HPA axis, and in the absence of any organic pathology is due to stress or burnout. You also have to look at the causes, which are the same for burnout, cfs and overtraining. --Sciencewatcher (talk) 16:02, 7 December 2007 (UTC)

You also have a look at the causes, "sciencewatcher". The officiall chorus in the scientific community is: "Cause unknown", you should learn to come to terms with that. If stress is listed anywhere as a theory - like on the CDC Website ( http://www.cdc.gov/cfs/cfscauses.htm )- it is listed as one possible cause among other possible causes.,JayEffage (talk) 11:27, 10 December 2007 (UTC)

Stress is the most significant trigger according to research - have a look at the stress and trauma section of this article. --Sciencewatcher (talk) 19:47, 10 December 2007 (UTC)

That is web2.0 circualar reasoning.

Who says it is the most significant trigger? You? The CDC certainly doesn't. NICE doesn't either. That was a nice question. Come up with an answer.JayEffage (talk) 21:48, 10 December 2007 (UTC)

Er, no, the references in the "stress and trauma" section do. Maybe you should actually read them. --Sciencewatcher (talk) 18:31, 11 December 2007 (UTC)

Where does it say "Stress is the most significant trigger", sciencewatcher? Also this section has at least to slashable references, thanks for making me aware of that. JayEffage (talk) 10:22, 12 December 2007 (UTC)

PMID 9201648. Stressful events in the year preceding CFS occurred in 85% of patients, which was higher than any other factor. --Sciencewatcher (talk) 17:16, 12 December 2007 (UTC)

Yes, and their conclusion is - lo and behold - that stress MAY play a role. That CFS is most likels MULTI-FACTORIAL. I guess the difference between those statements and your rethoric are your POV then? Or do you have anything that says "stress is the most signifcant trigger"? JayEffage (talk) 17:52, 12 December 2007 (UTC)

No, I'm not aware of any other studies that compared stress to other triggers. The percentages clearly show that stress is the most significant trigger. Yes, it is multi-factorial, but stress was shown to be the most significant. It may not agree with your POV, but it's there in black and white. Further discussion is pointless. --Sciencewatcher (talk) 19:16, 12 December 2007 (UTC)

It's not as simple as that. The strong findings in the study mentioned above were unusual and have not been replicated by similar studies, so we should review these other studies too. I'll post the full details during the weekend. - Tekaphor (talk) 14:55, 13 December 2007 (UTC)

A trigger is a short-time event. Stress is therefore not a trigger, but a catalyst. The trigger is the infection. Guido den Broeder (talk) 15:39, 13 December 2007 (UTC)

It may not agree with your POV, but it's there in black and white. Further discussion is pointless. Sciencewatcher

No, it is not there in unambiguous black and white, SW. That is misleading. Once again, if you are going to invoke stress as the (or even a) primary cause/trigger of ME/CFS, you will have to explain and take account of the following:

1. Why the HPA axis changes do not exist before onset, or even in the early phases of the disorder:

In summary, evidence suggests that, although there is HPA axis disturbance present in subjects with operationally defined CFS, it is not present before the onset of CFS or during the early phases of illness, but develops once the illness has taken a more chronic course. 'The HPA axis and the genesis of chronic fatigue syndrome', Cleare A. PMID 15036250

See below for a more detailed extract from this study [1].

It should also be pointed out that in the Infection and Stress section in Archive 3, you stated:

...all stresses affect the HPA axis... 14:56, 3 September 2007 (UTC)

Well, if that is the case, then where is the measurable pre-onset HPA change that indicates pre-onset stress of sufficient magnitude to induce a serious (allegedly stress caused) disorder? This is a very serious problem with your claim, and cannot be dismissed out of hand.

2. That a major and very recent review found that:

Various potential risk factors for the development of ME/CFS have been assessed but definitive evidence that appears meaningful for clinicians is lacking. 'Risk factors for chronic fatigue syndrome/myalgic encephalomyelitis: a systematic scoping review of multiple predictor studies.' Hempel S, et al. PMID: 17892624

3. The fact that retrospective studies which rely on subjective recall (particularly about subjective and ill-defined phenomena such as 'stress') are considered problematic and unreliable. For example, see PMID: 11258213, PMID: 11276550, & PMID: 11996606.

4. That studies using the Oxford or Fukuda selection criteria are seriously confounded by including patients who are suffering from chronic fatigue only, but not from ME/CFS proper (a very different diagnostic category from plain 'chronic fatigue', as has been pointed out before).

Just to confuse the issue further, that still leaves the problem of negative pre-onset HPA findings for patients who are only suffering chronic fatigue.

5. That in the Etiology section of the Overview of the Canadian guidelines (PDF link), it states:

Most patients enjoyed a healthy, active lifestyle prior to the onset of ME/CFS.

If you read the entire Etiology section, you will note they don't mention (psycho-social) stress. You will also note that the two studies you cite as the proof of your stress claim, PMID 9201648 & PMID 10367610, were both published in the 1990s, well before the Canadian guidelines (2003), so the Canadian authors would have been aware of them.

6. Also in the Infection and Stress section in Archive 3, in the same paragraph, you said:

In all the detailed case histories I have read, I have never seen any CFS patient who has not been in the midst of some severe psychological stressor.

An extraordinary claim. Got some solid evidence to back that with, besides these unspecified case histories the rest of us don't have access to? And how did you have access to them? Do you do clinical or research work in this field? How many case histories are you talking about? Were they all from the same source (indicating potential bias)? Were you involved in taking those case histories (also indicating potential bias)?

(I presume you are referring to the patients' pre-onset situation. Clearly most patients are under serious post-onset stress, which doesn't help matters, nobody is debating that. But it says nothing about primary cause of the disorder, only the secondary consequences.)

Taking into account the above points, the options you have here are to either:

a) demonstrate that serious stress doesn't always cause measurable HPA changes, and explain why it doesn't occur at all (pre-onset) in ME/CFS in particular, (and also why the post-onset HPA changes don't show up in a substantial fraction of ME/CFS patients); or

b) show why Cleare, and others, are wrong about the behaviour of the HPA in ME/CFS; or

c) accept that there is no objective evidence that stress is a (let alone 'the') major cause/trigger, and that there is substantial objective evidence against that proposition, and hence accept appropriate changes to your claim in the opening sentence of the Onset section in the main article that

The majority of CFS cases begin after a period of stress in the year preceding the illness....

Clearly your central thesis is that stress (or at least an abnormal stress response) is the major cause/trigger of ME/CFS. Serious problems with this thesis have been previously pointed out, and some time ago, and you still haven't dealt with them. You are right that further discussion about this claim is pointless while you continue to fail to deal with them.

If you don't address the problems with the misleading claim you made in the onset section about the causal/triggering role of stress (and you have had plenty of time to do so), then I will soon move for some changes (in early January).

••••••••••

[1] In the Stress and Depression section I started in late September, and which is now in Archive 3, Sciencewatcher asked for the relevant text from the full paper about how Cleare determined that HPA axis changes were not in play before onset, or early on in the course of the illness.

The full relevant text is further down, but the critical summary from Cleare is:

In addition, there is no evidence that an abnormal initial stress response is implicated. As it stands, the evidence suggests that during the early stages of fatigue genesis, the HPA axis is not an important factor. Instead, HPA axis changes probably develop somewhat later in the natural history of the illness. 'The HPA axis and the genesis of chronic fatigue syndrome', Cleare A. PMID 15036250

From the 'Prospective cohort studies' section:

One of the major risk factors that has been studied is that of severe viral illness [2]. For example, about six months after an episode of Epstein-Barr virus infection, ~22% of sufferers will have chronic fatigue and 9% more strictly defined CFS [15], in contrast to common and minor viral infections, where the rates of CFS are not higher than normal [16]. A recent study obtained salivary cortisol day curves to assess the HPA axis during an acute Epstein-Barr virus infection and at three- and six-month follow up [17]. Based on previous theories regarding the HPA axis and CFS [18,19], it was hypothesized that those who went on to develop CFS might show a more pronounced HPA axis reaction to the initial (viral) stressor. Furthermore, it was hypothesized that those who remained chronically fatigued at six months would show evidence of hypocortisolism if this were an important factor at that stage. The results showed that neither hypothesis could be supported: there was no difference in cortisol levels between those with chronic fatigue six months after the infection and those without. Similarly, the cortisol levels at the time of infection did not predict who remained fatigued six months later. Unfortunately, numbers were too small to test specifically whether those with strictly defined CFS, as opposed to chronic fatigue, at six months showed any HPA axis changes.

Another suggestion is that there may be some form of pre-existing risk factor or trait abnormality in the HPA axis that predisposes individuals to develop CFS, perhaps related to childhood abuse or chronic stress [20]. Studying individuals during an acute infection does not allow this to be assessed and other high-risk cohorts are needed where measures can be taken before and after a predictable trigger for fatigue. One example is operative stress [21]. One currently unpublished study from our group measured a daytime salivary cortisol profile one week preoperatively and on three occasions in the post-operative period up to six months later (G. James Rubin, PhD Thesis, University of London, 2003). We found no evidence of low cortisol either before surgery, or six months later, in those who were chronically fatigued six months postoperatively.

In summary, results from these two cohorts provide the first good prospective evidence that becoming fatigued during the first six months after an acute precipitant is not linked to underactivity of the HPA axis. In addition, there is no evidence that an abnormal initial stress response is implicated. As it stands, the evidence suggests that during the early stages of fatigue genesis, the HPA axis is not an important factor. Instead, HPA axis changes probably develop somewhat later in the natural history of the illness. Clearly, longer-term follow up of such cohorts, and larger samples to enable the identification of any possible subgroups of patients, are needed to confirm this suggestion.

References

2. Wessely, S. et al. (1998) Chronic Fatigue and its Syndromes, Oxford University Press

16. Wessely, S. et al. (1995) Post infectious fatigue: a prospective study in primary care. Lancet 345, 1333-1338

17. Candy, B. et al. (2003) Predictors of fatigue following the onset of infectious mononucleosis. Psychol. Med. 33, 847-855

18. Scott, L. V. et al. (1998) Blunted adrenocorticotropin and cortisol responses to corticotropin-releasing hormone stimulation in chronic fatigue syndrome. Acta Psychiatr. Scand. 97, 450-457

19. Scott, L. and Dinan, T. (1999) Neuroendocrine abnormalities in chronic fatigue syndrome: focus on the stress axis. Balliere's Clin. Psych. 3, 419-432

20. Heim, C. et al. (2000) The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology 25, 1-35

21. Rubin, G.J. and Hotopf, M. (2002) Systematic review and meta-analysis of interventions for postoperative fatigue. Br. J. Surg. 89, 971-984

Bricker (talk) 01:16, 14 December 2007 (UTC)

Bricker, thanks for taking the time to post this. I had asked for more details of the Cleare study in the past, and now you have provided it. I appreciate it.

Yes, you are right, this study does not look good for theories involving the HPA axis. Has this study been put in the article? If not, it should. Other studies also show differing results for cortisol/HPA axis activation in CFS patients (and in burnout), so it isn't really a reliable indicator for either condition.

I agree that asking retrospectively about stressful events is not the best way of doing things. I would have liked the Cleare study to ask patients about their stress levels after contracting EBV but before developing CFS. That would have been a more reliable measure of whether pre-existing stress when combined with the EBV infection made it more likely to develop CFS. More research is clearly needed. --Sciencewatcher (talk) 16:04, 14 December 2007 (UTC)

You're welcome. I appreciate that you are prepared to take on board this important evidence. I agree that HPA studies are generally not consistent, and that at best the results are only useful at a population research level, not for individual clinical diagnosis or treatment.

I have run out of time for now, and will get back here around the second week in January. In the meantime perhaps you (and others) could consider the evidence we have to date with an eye to revising the opening statement of Onset section.

I think the wording about the causal/etiological role of stress should be guided primarily by the Hempel, Viner, and Cleare studies (which are all major reviews).

Bricker (talk) 00:52, 18 December 2007 (UTC)

I posted the promised comments to the new "Stress and infection in CFS revisited" section of this talk page, albeit a week late! Perhaps the other recent comments above not relating to burnout could be moved there too. - Tekaphor (talk) 04:44, 22 December 2007 (UTC)

Alternative medicine

Guido, the Hong Kong paper has a PMID number and therefore is peer reviewed, a likely reliable source, it could be tested on the Reliable source noticeboard. Acupuncture certainly is no longer Fringe Theory (discussed by the mainstream) These Refs you might consider also discuss Alternative medicine and CFS. PMID 17459162 PMID 15889950 PMID 16444659 PMID 10882883 Jagra 05:14, 1 December 2007 (UTC)

Jagra, could you expand on those three PMID numbers, rather than expecting us to parse them ourselves?

The Hong Kong paper is a case report. I think case reports usually fall below the notability horizon. The orthomolecular thing needs more discussion, as the J Orthomol Med is obviously the organ of the orthomolecular movement. JFW | T@lk 23:00, 1 December 2007 (UTC)

JFW, The current Article wording says in effect, that CFS patients use alternative medicine partly because conventional treatments are poorly tolerated or ineffective. This study found certain non-prescribed supplements more effective than prescribed medicines and psychological support. Treatments at 6 months that predicted significant subsequent fatigue improvement were vitamins, exercise and Yoga. PMID 15889950 This study on CFS discordant twins found individuals with CFS frequently used alternative medical treatments and a large proportion found alternative therapies helpful; and more of those with CFS than those without CFS discussed their use of alternative medicine with a physician. PMID 10882883 Whilst the third study found people with CFS-like illness were significantly more likely to use body-based therapy (chiropractic and massage) and compared to non-fatigued persons, those with CFS-like illness or chronic fatigue were most likely to use body-based and mind-body therapies. All categories of CAM use were associated with significantly poorer physical health scores, but use of alternative medicine systems was not associated with significantly poorer mental health scores. PMID 17459162

So these 3 references support the general line of the wording. As to modalities I agree that needs further consideration. Unless I am wrong this Hong Kong study on acupuncture in CFS is a blinded controlled trial of 100 that supports the wording? PMID 17997936 It is further supported by these for acupuncture alone PMID 16318125 PMID 16248250 and by these refs of large study on combined treatments including acupuncture? PMID 17710799 PMID 17078435

The effective use of vitamins by patients is of interest as there have been calls for the use of anti-oxidants in CFS by researchers PMID 17693979 PMID 11703165 PMID 10767667 PMID 12531455 PMID 9367343 PMID 17239370 PMID 13677625 PMID 12639396 and by support groups. Trials have been carried out with injected GSH, PMID 11365019. Many of these substances could be classified as alternative medicine, so I will look further into specific references to cite.Jagra 04:54, 2 December 2007 (UTC)

Jagra, you are blinding me with science. You know what: let's do WP:BRD. You write something that is not WP:NOR or WP:SYNTH on the uses of alternative medicine in CFS. Also see my next comment on in vitro studies. JFW | T@lk 09:39, 2 December 2007 (UTC)

The "orthomolecular medicine" reference is poor science: it is fringe theory, and there were no controls. The acupuncture reference is almost as bad. If you look at the actual results, there was not a very significant difference between the controls and active group. The controls symptoms were reduced just slightly less than the active group. Also, they did not ask patients to guess which treatment group they were in, so there is no way of determining whether the patients were actually blinded. Most likely they were not, because the sham treatment did not penetrate the skin. --Sciencewatcher 05:39, 4 December 2007 (UTC)

I agree that one single reference with a poor-quality study may not be suitable for inclusion. I am already opposed to the single case report on acupuncture. Sciencewatcher, you must agree that several studies have shown that patients cannot tell the difference between sham and real acupuncture needles; sham acupuncture is the only way to perform randomised double-blinded controlled trials with real acupuncture.

I am still waiting for Jagra's next move in the BRD cycle. JFW | T@lk 05:52, 4 December 2007 (UTC)

Thanks. The text should indicate that such treatments are offered, not that they are effective. I have changed the wording, hopefully it is more clear now. Guido den Broeder 10:48, 4 December 2007 (UTC)

But apart from a single case report, do you have any other evidence that acupuncture is "offered"? One single case in a Chinese journal (and therefore not easily verifiable) is absolutely insufficient to serve as a basis for such a generalisation. JFW | T@lk 11:34, 4 December 2007 (UTC)

JFW: the problem is that as far as I am aware, there is no acupuncture sham which actually works (i.e. where the patients do not know whether or not they had the real treatment). For the study to have any value, they need to test the blinding by asking patients to guess which treatment they had. This acupuncture trial did not do that. Currently the only sham procedure for acupuncture that has been shown to work is to actually puncture the skin, but to do it at a non-acupuncture point. When this is done there is no difference between acupuncture and the sham procedure, meaning that acupuncture is either completely placebo effect, or that it is the needles puncturing the skin that causes the effect (but it doesn't matter where you stick them). Overall, acupuncture research (and in fact most CAM research) is riddled with shoddy trials which do not have effective sham procedures. --Sciencewatcher 16:23, 4 December 2007 (UTC)

I have added a second article on acupuncture. The fact that an article is in Chinese does not make it less verifiable. That's your shortcoming; not the article's. Guido den Broeder 16:31, 4 December 2007 (UTC)

The article you added had no controls, so I removed it along with the others you also added. Please stop adding worthless studies to support your position. --Sciencewatcher 16:35, 4 December 2007 (UTC)

Please refrain from editwarring and insults. The text is, by the way, unrelated to 'my position', I merely improved and sourced an older text. Guido den Broeder 16:50, 4 December 2007 (UTC)

All the alternative medicine section is saying is that there are modalities being researched. Really, only acupuncture is being researched. The orthomolecular medicine reference is to a single case study, of a single patient, nearly eight years ago. That means that orthomolecular medicine was reported once, to work for a single person. That's not a program of research, that's a single case report. I think given the statement in the article, it's best to call that section 'acupuncture' and leave out alternative medicine barring other sources. WLU 21:09, 4 December 2007 (UTC)

Agreed, for the moment. If we find other research, we can always expand. Thanks, Guido den Broeder (talk) 22:30, 4 December 2007 (UTC)

There are lots of speculative treatments being researched, but they should not be listed in an encyclopedia article unless they have been shown to work. The references given for acupunture do not show that it helps cfs at all. --Sciencewatcher (talk) 21:56, 4 December 2007 (UTC)

Given the continuous back-and-forth reversion, I have now requested protection (diff) so we can actually work out a consensus rather than be silly. Jagra has posted a large number of studies, but I have no access to most of them and would much prefer if any contributor (preferably Jagra himself) could propose a non-controversial paragraph here that summarises alternative medical approaches being studied for CFS. Keeping in mind my comment below, the studies need to be clinical studies with clearly set clinical endpoints, rather than looking at CD56+/CD16- lymphocyte subsets or other surrogate markers. JFW | T@lk 02:32, 5 December 2007 (UTC)

I haven't been on-line much the past three days, so I haven't seen it. Right now the page reads Acupuncture is being researched as an alternative treatment for chronic fatigue syndrome, with some evidence suggesting that acupuncture may reduce symptoms.[259][260] Other alternative treatments are sometimes proposed for CFS, especially when conventional treatments are poorly tolerated or fail to relieve symptoms. Seeing as the two articles referenced are pubmed articles, I think that's a reasonable summary of the two articles. The other sentence doesn't have references, but they do seem reasonable if no-one else objects. That's my two cents, I'm happy to see this version protected and I don't think much else could be said given the references. WLU (talk) 02:35, 5 December 2007 (UTC)

I'm with WLU on this one. It goes against Wikipedia to assume that only English sources are verifiable. And constantly deleting a section while others are working to improve it does not strike me as constructive. Guido den Broeder (talk) 09:17, 5 December 2007 (UTC)

Despite Guido's insinuations, I have my doubts about the verifiability of non-English references. The first one of the Chinese studies has a control group, which used sham acupuncture. Contrary to Sciencewatcher, I believe that sham acupunture methods have been adequately validated (example). If it would be possible to find an English reference supporting claims about acupuncture research, that would be much to be preferred.

Obviously many more alternative modalities "are being investigated" for CFS. Tiredness is a very common presentation where conventional medicine has little to offer, and where complimentary/alternative medicine may offer detoxification, healing, meditation, crystals, dolphins and other things more. I cannot believe that nobody has documented efforts to apply these to "case definition positive" CFS (or even ME). JFW | T@lk 03:02, 5 December 2007 (UTC)

JFW, please do not confound the fatigue in CFS with normal tiredness (or even with 'chronic fatigue'), the distinction between them is fundamental, and not just a matter of degree or semantics. This issue has been thrashed to death in the scientific debate and that critical distinction is now widely accepted by mainstream researchers and clinicians, for example the Overview of the Canadian Guidelines (PDF link);

“Fatigue” is an inappropriate label because the fatigue experienced in ME/CFS is not normal fatigue whereby energy is promptly restored with rest. The pathological “fatigue” experienced in ME/CFS may combine exhaustion, weakness, heaviness, general malaise, lightheadedness, and sleepiness that can be overwhelmingly debilitating.

and the Fukuda 1994 definition;

...the symptom of fatigue refers to severe mental and physical exhaustion, which differs from somnolence....

Bricker (talk) 01:51, 14 December 2007 (UTC)

JFW: I had a look at the article about sham procedures you posted, and it does seem to be valid. But what procedure did this Chinese study use? Unless someone can answer that simple question, the study has to go... --Sciencewatcher (talk) 03:13, 5 December 2007 (UTC)

Also: have a look at Table 1 (inside the full text article). It seems pretty clear that the results are NOT significant (even though the authors claim they are). Example: the difference between the "before" scores of the control and active groups is twice as high as the difference between the "after" scores! Also the control group's reduction in symptoms is about 80% of the active group, meaning that the placebo effect is accounting for at least 80% of the improvement. And the difference in scores between active and controls is about 25% of the standard deviation. Any way you look at it, it is an incredibly poor study and it certainly doesn't give any credence whatsoever to the idea that acupuncture helps CFS. --Sciencewatcher (talk) 03:30, 5 December 2007 (UTC)

The full text of this article can be found here. Reduction of fatigue symptoms in the control group is 53% (mental) and 65% (physical) of that of the trial group, not 80% as Sciencewatcher claims. Improvement in SF-12 score was 54% of that of the trial group. There is no reason to assume that these differences are wrongly reported as significant, Sciencewatcher is comparing apples (standard deviations within a group) with pears (standard deviations of changes). Guido den Broeder (talk) 10:05, 5 December 2007 (UTC)

The reduction in symptoms is meaningless when the baseline score was so different between the control and active groups. If you take 30 as the baseline for the physical fatigue before score, then the reduction of the control vs active group is 81.7%, as I said. And how can a difference of 1.17 in the mental fatigue after scores be "P<0.01" when the difference in baseline scores between the two groups was 2.1 before treatment! Please explain to us all how you think the figures were fudged to calculate the P<0.01 so we can all have a good laugh. --Sciencewatcher (talk) 16:32, 5 December 2007 (UTC)

As for your 81,7% - the baseline wasn't 30 in both groups, was it? JayEffage (talk) 19:21, 5 December 2007 (UTC)

It was different in both groups, and that is my point. Obviously the group with the higher symptoms before the study is going to improve more. It is just as valid (or invalid) to compare the final outcome against a common baseline than to compare against two different baselines. --Sciencewatcher (talk) 20:42, 5 December 2007 (UTC)

"Obviously the group with the higher symptoms before the study is going to improve more". Could you explain that statement a bit please?

Explain please why the improvment here is not only bigger in relative but also in absolute terms (After treatment value for mental fatigue in sicker group better than after treatment value in sham-treatment group)JayEffage (talk) 21:26, 5 December 2007 (UTC)

I can think of a number of possible reasons. If the treatment cures the patient, then the original baseline is irrelevant. If a patient is sicker at the start then he will obviously have a greater improvement when cured. But more to the point: how can the result be significant if the original groups differed by more at the start than at the end? The study is assuming that the difference at baseline is non-significant. Therefore, logically, the (smaller) difference between the groups after treatment must also be non-significant. --Sciencewatcher (talk) 22:24, 5 December 2007 (UTC)

You missed my point. You are explaining a general, hypothetical scenario in which patients are cured (not the case in this study) and the relative gain in health by the group which was sicker before it was cured is trivially bigger than the gain in health by the group which was less sick before it was cured. In this study, the patients are still sick. But the group with the sicker patients before treatment is now the group with the less sick group in ABSOLUTE terms. (Or in other words: The sicker group is cured beyong the point of cure of the less sick group.) The algebraic signs of the two differences are not equal, they are opposite. So the study must not necessarily assume that the difference at baseline is non-significant. JayEffage (talk) 22:56, 5 December 2007 (UTC)

I think you missed my point. The whole point of a control group is that it is the same population as the treatment group. If you have to assume that the control group is statistically significantly different from the treatment group at baseline then the study is meaningless. Error and bias works in both directions, and that is more likely to explain the results than the treatment. --Sciencewatcher (talk) 01:14, 6 December 2007 (UTC)

I think it is the original authors who were comparing apples and pears. If you use their logic then the control and active groups also fit the P<0.01 significance for being different, rendering the entire trial useless. Can someone shed some light on the exact calculation that is used to determine these probabilities? --Sciencewatcher (talk) 16:38, 5 December 2007 (UTC)

You might gain some insight if you first try and formulate what these probabilities stand for (these are probabilities of what?). Hint: they have nothing to do with what you have said so far. Guido den Broeder (talk) 17:09, 5 December 2007 (UTC)

Your smugness is unbecoming and though not uncivil, isn't helping with the acrimony on the page; if you've a problem with SW's conclusions, state them. We're not children, and there is some merit in my mind to SW's objections. Sciencewatcher, my problem with this edit is that it is verifiable that acupuncture has been investigated and as reported in a reliable source, there's some evidence it might work. Non-english sources are acceptable, and with a pubmed number I'm inclined to say it's a reliable source. It's phrased as a suggestion of symptom reduction, not a definitive treatment. I agree that the stats don't look great, and their conclusions are wildly inappropriate, but the articles exist and do add information to the page pending better sources. Perhaps we could agree on a watered down version of the current page, even just 'acupuncture is being investigated as a symptomatic treatment for CFS'? WLU (talk) 18:49, 5 December 2007 (UTC)

Hey Guido, I know exactly what the probabilities mean. I just don't know off the top of my head how to calculate them. But it is clear that the numbers they have calculated have no basis in fact - they have either misunderstood or <deleted>, and you are doing the same. And you haven't answered any of my criticisms. I am not going to waste my time figuring out how to correctly calculate the probabilities for an obviously poor study. If you really want to defend the study, perhaps you can post an analysis showing how they might have arrived at those probabilities, so that we can critique it.

WLU: Having a pubmed number is definitely not a guarantee of being a reliable source. --Sciencewatcher (talk) 20:36, 5 December 2007 (UTC) deleted insult Guido den Broeder (talk) 23:27, 5 December 2007 (UTC)

@everybody: If you run the paper through babelfish, you can see that they used "variance analysis". That's the only mehtod name I found in the gibberish. if you run the numbers (active and control for mental fatigue) through a Forumula for "F" and compare to a fisher's table for 0.01 you will find that it it holds.JayEffage (talk) 21:26, 5 December 2007 (UTC)

Hardly worth the trouble though, nobody in their right senses - be them western or eastern - will try to cheat you in statistics by rigging anything in a traceable (e.g. numerical) way. The devil is mostly in the things that conveniently don't get a mention.

I think we should settle for WLU's suggestion and leave fisk and counterfisk to the friendly folks at the accupuncture page. JayEffage (talk) 22:10, 5 December 2007 (UTC)

If you do a 2-tailed student's t test on the baseline scores for mental fatigue for the 2 groups, you get a probability of 0.0424, which is statistically significant. And if you do the test on the "after" scores, the probability is 0.22, which is not statistically significant. Therefore the control and treatment groups are statistically different BEFORE the treatment, but not after. I think the problem is that the study is using meaningless statistics. --Sciencewatcher (talk) 23:25, 5 December 2007 (UTC)

These are irrelevant comparisons. The hypothesis is that the trial group improves more than the control group. That, and only that, is what is the p-values reflect. Guido den Broeder (talk) 23:36, 5 December 2007 (UTC)

But that statistic is meaningless if the trial and control groups are statistically significantly different at baseline (which they are). The populations are more different at baseline than after the treatment. The whole point of a "control" group is that it comes from the same population as the treatment group, and the study variables should be similar in both groups at baseline, otherwise the entire study is completely meaningless. The same errors and bias that result in differences at baseline will result in differences after treatment. --Sciencewatcher (talk) 01:14, 6 December 2007 (UTC)

Although the match could have been better, the two groups do not seem to be significantly different from each other before treatment. Guido den Broeder (talk) 01:35, 6 December 2007 (UTC)

See my comment above regarding the Student t test, which confirms they are significantly different before treatment. That is the reason why the study is invalid (along with not validating the blinding). You can confirm the t-test yourself here--Sciencewatcher (talk) 03:36, 6 December 2007 (UTC)

You cannot interpret your outcome that way. The groups have four characteristics. Odds are, that such a test will show a difference in one of them purely by chance. Taken together, the four differences are within normal range. Guido den Broeder (talk) 11:20, 6 December 2007 (UTC)

Well, this discussion has nothing to do with WLU's suggestion anymore. For what it's worth, I did the t-test as sciencewatcher suggested - and it turned out as he claimed(= not significant). If, however, you do a variance analysis as the authors of the paper did, you do indeed get a significant result. That is odd, one good explanation is that I did it wrong - but then the authors would have done it wrong as well, which I doubt. In addition, it raises the question why they used that method.

I personally disagree with sciencewatchers conclusion that the ex-ante difference in active and controll group renders the study invalid, because the direction of the difference changes before and after the study - that is an extra piece of information which cannot be discounted on a purely rethorical basis like "there was a difference - therefore invalid". Although it gives the entire paper a strange taste.

If we want a definitive answer, we have to contact the authors (email is in the pdf...) and as them about their methodology, if the control/active differences are accident or purpose etc..

Alltogether, it's probably not worth the effort. I still think we should go for WLU's suggestion.JayEffage (talk) 11:50, 6 December 2007 (UTC) (sorry, forgot to sign)11:48, 6 December 2007 (UTC)

I'm sorry, but SW's t-test is invalid, so yes, you did it wrong, too, but the authors didn't. You cannot look at the outcomes and then say: "Hey, this difference seems big, let's test it separately." The proper test is the probability of the largest of four differences being that big or bigger. Another one is to test the pattern as a whole. These are tests that we cannot perform, because we don't have the covariances, but rest assured that they will come out OK. A rough estimate for the second test would be to take the geometric mean of all four p-values. Guido den Broeder (talk) 12:49, 6 December 2007 (UTC)

Perhaps you will understand it this way: the p-value for a separate test on the second largest difference, physical fatigue, is 0.1647. This means, that SW is investigating a p-value of 0.0424, for something that he picked to be less than 0.1647 for certain. The t-test, however, knows nothing about SW's deliberate choice to select the largest of four differences, and merely assumes that it is less than 1, i.e. it assumes a full t-distribution, while in reality it is truncated: a large portion of all the probabilities is unavailable. Therefore, the true p-value for SW's test is 0.0424/0.1647=0.2574, which is not significant. Regards, Guido den Broeder (talk) 13:52, 6 December 2007 (UTC)

But Guido, why should SW's test be invalid?

And if I did it wrong, then why did I reach at the same conclusion as the Authors, namely the difference after treatment for mental fatigue having p < 0,01? Please note that the authors used a little symbol (two delta signs/triangles) to denote the p < 0,01 signigicance specifically for the outcome of mental fatigue. Or am I reading that wrong?

You might still have a point, because the text talks about some sort of repeat analysis (and that's where variance analysis is commonly used). As I said, if you want a comprehensive answer, you will have to contact the authors, we can just guess what they did or didn't do.

If you think that's worth it...I am perfectly happy with WLU's statement.JayEffage (talk) 14:09, 6 December 2007 (UTC)

You have a test with 19 white balls and one red ball. Then you remove 16 of the white balls from the test without telling anyone, and still claim that if the red ball comes out something must be terribly wrong. Guido den Broeder (talk) 16:20, 6 December 2007 (UTC)

Been off-line for 4 days for technical reasons, unable to respond to JFW invitation, seems things have moved on, so I will add to discussion whilst protected. Support WLU comment and think the sentence "Other alternative treatments are sometimes proposed for CFS, especially when conventional treatments are poorly tolerated or fail to relieve symptoms" is supported by the three references commented on at the top ie. PMID 17459162 PMID 15889950 PMID 10882883. Except that I would phrase it as "CFS patients use alternative medicine, at times in consultation with their practitioner, partly because conventional treatments are poorly tolerated or ineffective".When it comes to modalities there is some evidence also for anti-oxidant use in CFS, and I will add this further.Jagra (talk) 05:09, 6 December 2007 (UTC)

Guido: you don't seem to have understood what I'm talking about. You mentioned "differences" but that is not what I was testing (and I didn't have the information to do that anyway). The Student t test is for comparing two groups to see if they came from the same population. That is what I was testing. The control and treatment groups at baseline are significantly different according to the t test. We have all the information that is needed to do that test.

Jay: surely it doesn't matter that the signs are different. Think of it this simpler way: there is a large error/bias/noise/whatever in the "before" scores. The difference in the "after" scores is actually less than this noise, and is therefore irrelevant. Of course you can come up with statistics that will give a "significant" result, but statistics must be based on some kind of foundation to be meaningful and valid. --Sciencewatcher (talk) 15:51, 6 December 2007 (UTC)

If you want to test whether the groups are different you must test on all four variables simultaneously. This t-test is only for the simple case where there is just one variable.

The hypothesis is not that the treatment creates a difference between the two groups, but that the improvement of the trial group is larger than of the controls. Guido den Broeder (talk) 16:29, 6 December 2007 (UTC)

As I said, for the study to be valid the variables at baseline have to be similar between the two groups. You can do all the statistics you like, but the study is still invalid. You can use the t test to test each individual variable at baseline and you'll see that they are significantly different. That is what the t test is designed for, and that is how you use it. You certainly do not need to "test all 4 variables simultaneously". --Sciencewatcher (talk) 16:39, 6 December 2007 (UTC)

Yes, you do. If the rules of statistical inference were as you claim them to be, you could render any two groups different by simply adding more variables until one pops out that fails your t-test. Say, 8 patients in the first group were born on a Monday, and only 6 in the second group. Then, according to you, the study would become invalid when we added these data. Guido den Broeder (talk) 16:48, 6 December 2007 (UTC)

All that this t-test does is estimate the probability that a single-variable difference is a random occurrence. It is not a proof; the cut-off point is arbitrary. We simply agree that if the probablity is less than e.g. 5%, it's too unlikely. But we could be wrong, it could still be a coincidence. Now, if we have more variables, the odds of finding one such difference purely by chance increase dramatically, to {1 - 95%x95%x...} if they are uncorrelated. Therefore, we won't use the same cut-off point. Get it now? Guido den Broeder (talk) 17:12, 6 December 2007 (UTC)

But the study results are each calculated on single variables (mental fatigue, physical fatigue, etc.) so if it's valid for the study to base its results in individual variables, then it is equally valid to do the same for the baseline scores. --Sciencewatcher (talk) 20:49, 6 December 2007 (UTC)

No, it's not. Furthermore, that's not what you set out to do, not what you did, not what you claimed to do. Guido den Broeder (talk) 21:37, 6 December 2007 (UTC)

So, ignoring Guido's last incomprehensible comment, can we assume that nobody has any objections to deleting this acupuncture reference due to the severe methodological problems? I suggest we revert either to my last revision, or to OrangeMarlin's revision on 30 November (where the alternative medicine section is simply deleted). While there is evidence that alternative medicine helps CFS, there is no evidence that it is any better than placebo. As for acupuncture, there is good evidence that poking needles into the skin reduces pain over and above the placebo effect, but it doesn't seem to matter whether you poke them into acupuncture points or non-acupuncture points. So it might in theory reduce pain in CFS, but I can't see it helping with much else. --Sciencewatcher (talk) 16:55, 8 December 2007 (UTC)

I object as well. Guido's last statement is neither less nor more valueable as your above non-backed-up drivel about your personal feelings regarding alternative medicine and what you feel about the placebo effect. The statistical question is not settled - and while the discussion was sure interessting and partly revealing - this page is not the place to pick the contents of a study appart.

Again, I would go for WLU's suggestion, that accupuncture is researched in CFS. JayEffage (talk) 11:15, 10 December 2007 (UTC)

Well you can either have a proper discussion without resorting to personal attacks, or we will just ignore your comments and proceed regardless. Your choice. If you want a reference for anything, just ask nicely and I will be happy to give it. I don't respond to personal attacks, and they are not appropriate here. --Sciencewatcher (talk) 19:44, 10 December 2007 (UTC)

No, sciencewatcher, YOU don't procedd, because you don't have consensus.

If you don't want your statements to be called drivel back them up with evidence otherwise they are clinically inseperable from drivel sensu stricto.

As for your offer to answer questions which have been asked nicely, I suggest you finally fulfill this promise you have given so many times and come up with answers for the other questionable claims like "there are lot of autopsies", "CFS = Burnout" before we ask new questions.JayEffage (talk) 21:41, 10 December 2007 (UTC)

The rules of statistical inference are what they are, they don't change just because SW doesn't know them. With regard to autopsies: Dr. Peterson tried to get them done at Lake Tahoe back in the 1980's, but nobody wanted to do any. He still has many tissue samples waiting for analysis. The situation has not changed; autopsies on ME/CFS patients exist but they are very rare. Guido den Broeder (talk) 12:34, 11 December 2007 (UTC)

Guido, "mental fatigue" is a single variable. The result we are talking about is "mental fatigue after - control vs treatment" (the most significant result of this study). Your argument about multiple variables is irrelevant. It is clear that the initial baseline scores differ significantly for mental fatigue - you can tell this just by eyeballing the results, or you can use statistics to confirm it. --Sciencewatcher (talk) 17:55, 11 December 2007 (UTC)

You are cheating by ignoring the regularity conditions for this t-test. Guido den Broeder (talk) 20:44, 11 December 2007 (UTC)

Please explain. I wasn't aware of any "regularity conditions". The t test measures two samples to see if they are from the same population - nothing more, nothing less. You can use it in this case, or you can just eyeball the data. Either way it is obvious that the two groups have significantly different baselines. This is getting frustrating - you just keep coming up with more and more nonsensical excuses. --Sciencewatcher (talk) 21:35, 11 December 2007 (UTC)

I understand that it is frustrating. If my earlier explanation and the terms that I use don't make sense to you, then you simply lack the appropriate basic knowledge to build on. There is no quick way to teach you this stuff; it took me 6 years to learn it. Guido den Broeder (talk) 22:28, 11 December 2007 (UTC)

Again you resort to smug, condescending remarks with no actual content. Please explain or we will assume you just made that up about "regularity conditions". Also do not accuse me of vandalism when it is patently obvious that I did no such thing. --Sciencewatcher (talk) 23:13, 11 December 2007 (UTC)

Goodbye, Sciencewatcher. Guido den Broeder (talk) 09:26, 12 December 2007 (UTC)

Well I for one would object and add a further modality for discussion.

Alternative medicine and Anti-oxidants

A hyper-responsive immune system with resulting lipid peroxidation is associated with oxidative stress in CFS. Although there are several enzyme systems within the body that scavenge free radicals, the principle antioxidants are vitamin E, vitamin C, n-acetylcysteine and a-lipoic acid. The vitamins C and E are thought to protect the body against the destructive effects of free radicals. Oral administration with selected antioxidants, a-tocopherol (vit E), and ubiquinone (CoQ10) potentiate antioxidant defence systems. Additionally, selenium, a trace metal that is required for proper function of one of the body's antioxidant enzyme systems, glutathione (GSH). PMID 17535753 Zinc and L-carnitine are both anti-oxidant and increase glutathione. PMID 16338007 PMID 17826175 (see discussion item 10.3 above for more inter relationships) Researchers are calling for the use of anti-oxidants including these substances and others in CFS (see references above)

Research has found low levels of anti-oxidants indicative of high demand in CFS, including; ubiquinone (CoQ10), glutathione, and free-radical scavenging enzymes, (SOD and catalase) PMID 12745627 PMID 10872900 low vitamin E PMID 10872900 PMID 12531455 low serum Zinc PMID 16338007 lower serum total carnitine, free carnitine and acylcarnitine levels, PMID 12414265 PMID 9854142 PMID 8544970 Increased ascorbic acid (vitamin C) demand PMID 15715687

The following is proposed for addition to the Alternative medicine section of the Article; Supplementation/treatment with anti-oxidants, glutathione, ubiquinone (CoQ10), oral L-carnitine, acetylcarnitine, and propionylcarnitine, and vitamin C infusion, has shown benefit in CFS PMID 11365019 PMID 8241699 PMID 9018019 PMID 15039515 PMID 8986468 Serum vitamin E improved after magnesium supplementation PMID 10872900 Jagra (talk) 03:41, 10 December 2007 (UTC)

Jagra, all of these studies are again 'fringe science', and there are no placebo controls on the clinical trials, which makes them worthless. --Sciencewatcher (talk) 20:02, 10 December 2007 (UTC)

Less valuable, but not worthless. There are more carnitine studies, too, as well as information from practice and experience. Acetylcarnitine is one of the few medicines that has a positive effect for a significant percentage of the patients. Guido den Broeder (talk) 12:34, 11 December 2007 (UTC)

Sciencewatcher, so all of these are fringe eh! Sounds like a plaintiff call made when you can’t think of anything else. Where have I heard that before, that’s right you will remember, some time ago another editor levelled that accusation concerning Allostatic Load, I had that checked out then and it was shown to be a matter of POV. Let’s see how you fare, all of them eh? Given the apparent confusion let’s start with definitions;

Fringe science is, by definition, at the fringes of a mainstream academic discipline. Fringe science is a phrase used to describe scientific inquiry in an established field of study that departs significantly from mainstream or orthodox theories. Traditionally, the term "fringe science" is used to describe unusual theories and models of discovery that have their basis in established scientific principle. Mainstream here refers to ideas which are accepted or at least somewhat discussed as being plausible within major publications or respected and peer-reviewed academic publications.

In order to be notable, (enough for inclusion in Wiki) a non-mainstream theory should be referenced extensively, and in a serious manner, in at least one major mainstream publication or by another important mainstream group or individual. Even debunking or disparaging references are adequate, as they establish the notability of the theory outside of its group of adherents’

So all I have to do is to find the subject matter in a respecterd journal, to disprove your claim of fringe, and even if not, provided it is an RS verifiable source it can still be included. A search of Pubmed will show the following;

Carnitine and treat yeilds 1,918 RS papers

Ubiquinone and treat yields 946 RS papers

Glutathione and treat yields 8,910 RS papers

Vitamin C and treat yeilds 6,540 RS papers

Vitamin E and treat yields 7,794 RS papers

Magnesium and treat yields 10,060 RS papers

Put as simply as I can You have not established by a long shot that the use of any of these substances in treatments is Fringe and your call begins to look like POV to me. Jagra (talk) 02:41, 13 December 2007 (UTC)

You must have had a bad day Sciencewatcher! either that or you did not read the papers? The cited trial PMID 15039515 on carnitine esters did have placebo control (See free text) , which makes your comments of little substance Jagra (talk) 07:48, 13 December 2007 (UTC)

Jagra, I never said these substances were fringe science. Their use in treating CFS is fringe science. And the placebo trial only involved 6 patients in each arm, so while interesting it isn't very conclusive. --Sciencewatcher (talk) 16:19, 13 December 2007 (UTC)

So the use of these substances in treatment is not fringe science, only if used in CFS? It seems some at the mainstream UCLA School of Medicine, California, USA would disagree PMID 10767667, as would publication in the mainstream journal Ann NY Acad Sci PMID 15591010 as well the publication of the carnitine ester study in CFS was in the mainstream journal 'Psychosom Med.' and the CoQ10 study in 'Clin Investig' making your comment a matter of opinion. This would suggest your comment then is irrelevant, as on your ctiteria there would be few if any left standing. Or put another way use of such substances, found deficient in CFS and called for by researchers of CFS and found useful by CFS patients some that have been trialled and clinically shown to be useful is no more fringe than other questionable treatment already included in the Article. Even if this were so, this does not exclude the inclusion of such treatments from the Article suitably qualified. Jagra (talk) 04:24, 16 December 2007 (UTC)

Actually with the carnitine ester studies there were 6 placebo patients in each arm of the unpublished studies 'a preliminary open label study, that treated 150 CFS patients Another 18 CFS patients were included in a randomized double-blind study. Six were treated with oral acetyl-L-carnitine 1 g/d plus L-carnitine 1 g/d (low dose), 6 received twice the dosage (high dose), and 6 received placebos. (unpublished results).' In the reported study there were blocks of 6 placebo patients assigned to each of the 3 arms in the study. Jagra (talk) 10:16, 16 December 2007 (UTC)

Sorry, 'fringe science' was perhaps not a correct description. Let's just say it is unproven. As other people have pointed out, we cannot put all possible treatments into the article. --Sciencewatcher (talk) 17:02, 16 December 2007 (UTC)

Apology accepted, but 'unproven' is not quite correct given the carnitine studies. The unpublished studies were carried out to determine the parameters (including that for placebo's) for the cited study therefore its conclusions are more than just interesting or notable. 'This study indicates that (carnitine esters) are successful in the treatment of symptoms of CFS in a major subset of CFS patients.' Something few if any other treatments can claim, and therefore worthy of inclusion in the Article.Jagra (talk) 11:16, 18 December 2007 (UTC)

The study does not support the conclusion they reach. You cannot claim the treatment is successful if you only have 6 patients in each arm of the study, and it is therefore not worthy of inclusion. I think they are erroneously basing their conclusion on the results of the uncontrolled 150-patient trial. --Sciencewatcher (talk) 15:43, 18 December 2007 (UTC)

You seem confused Sciencewatcher, the 18 patients is another unpublished trial. The published trial has 90 patients, 30 in each arm, and therefore worthy. Jagra (talk) 11:13, 20 December 2007 (UTC)

No, I think it is you who is confused. The 90 patient trial had no placebo, and if you read their conclusion they actually say "Conclusive evidence for the effects of carnitine esters on CFS will have to await a double blind, placebo-controlled, randomized trial.". --Sciencewatcher (talk) 16:05, 20 December 2007 (UTC)

We agree with their conclusion, and intent of further trials, but not with your statement "and there are no placebo controls on the clinical trials which makes them worthless" Jagra (talk) 03:37, 21 December 2007 (UTC)

This article is about people

At the moment, we quote a lot of research that is primarily in vitro work. PMID 16648791, which is the only reference supporting a role for "dental amalgam", is primarily a laboratory study on lymphocytes, with clinical improvement added as some sort of afterthought. There is an enormous loss to follow-up (less then half of the original cohort), very much indicating that the study is open to recall bias - perhaps those who didn't improve couldn't be bothered to have their veins stabbed again and declined to respond. The abstract also doesn't indicate how improvement was measured - subjectively or through well-validated tools.

Ideally we should have some sort of standard on the quality of studies we base this article on. JFW | T@lk 09:39, 2 December 2007 (UTC)

I think that's why we're here as editors. One or two people can throw in a reference, without actually reading it, and claim it proves one point or another. A neutral point of view requires reliable sources, avoidance giving undue weight to fringe theories, and not creating a conclusion when one isn't there. I read these references put in by alternative supporters--the strongly positive ones usually come from non-peer-reviewed sources and are biased, or they actually don't support the statement. Either way, we should delete. OrangeMarlin ^{Talk• Contributions} 02:39, 5 December 2007 (UTC)

Seems to me the Melisa test is merely pathology and is used as such,in this study, test- diagnose- treat- test again- observe (people). Tell me why it should be considered otherwise? There are many reasons testing is done 'invitro' and not 'in-vivo' including for ethical reasons. Still a large study on 248 people!Jagra (talk) 04:45, 6 December 2007 (UTC)

There are many problems with this study. What is the mechanism for these metals causing CFS when they aren't known to do that? Where is the control group? Where is the randomised placebo control (and how do you do that?) Bear in mind that the placebo effect can result in significant changes to the immune system which could cause the Melisa results, and dental amalgam removal will obviously generate a huge placebo effect. As Orangemarlin says, this study simply doesn't support the author's conclusions. It is poor, fringe science. --Sciencewatcher (talk) 16:46, 6 December 2007 (UTC)

Sciencewatcher, you too are missing the point, Toxic agents comes under "PROPOSED CAUSES and PATHOPHYSIOLOGY" and the Article wording says '(may) have a possible effect on the cause and/or course of CFS'. The authors propose an improvement on the course, no one has said it is the cause! We do not know how long the follow up period was, but the reason for such longer follow up is because placebo effects wear off quickly, if as Bricker points out below there were any at all, 'a huge placebo effect' is your POV? Given I have already posted 3 RS papers by seperate authors with similar results and Pubmed shows 18 studies in a range of journals, I think it easy to demonstrate it is discussed by the mainstream and not likely 'fringe'. What is the basis for your assertation, show me something RS that says is fringe. For your edification Orangemartin did not make any such comment in relation to the paper under discussion, but to this one PMID 11462117 a smaller study. Jagra (talk) 05:59, 7 December 2007 (UTC)

The kind of placebo effect that would be in effect here would be the same as for depression, which gradually takes effect over a period of months. I suggest you do some more research. I wasn't giving a POV, just a possible explanation. The placebo effect will be huge if the circumstances are right, as they are in this experiment, whether or not there is a clinical effect. It doesn't matter how many papers you have if they all have the same methodological problems. There are hundreds of papers "proving" that homeopathy works. --Sciencewatcher (talk) 15:41, 7 December 2007 (UTC)

How can you have a "huge" placebo effect, but no clinical effect? Bricker (talk) 04:31, 8 December 2007 (UTC)

I meant no effect from the treatment. Why don't you do something useful instead of always insulting and picking hairs, like actually contribute to the article? --Sciencewatcher (talk) 16:25, 8 December 2007 (UTC)

While of course, the extensive statistical analysis you just took part in on this talk page, wasn't picking hairs. Give us a break, sciencewatcher...JayEffage (talk) 16:47, 8 December 2007 (UTC)

So we agree it is not 'fringe', but you do not think the treatment might have a 'possible effect on the course of CFS' because of a supposed placebo effect the authors had not considered. How about you do your research and post some references to prove your point of the 'huge placebo effect'Jagra (talk) 03:51, 10 December 2007 (UTC)

It certainly is 'fringe', and I think it is unlikely to have any effect. Have a look at the wikipedia entry for the placebo effect which gives all the references you need. It isn't necessary to prove that the placebo effect is causing the effect in this case. There could be many other possible explanations, such as natural course. It would be impossible to prove any of these without doing the experiment again using controls. --Sciencewatcher (talk) 19:38, 10 December 2007 (UTC)

Well, it seems large numbers less likely placebo was factor. The experiment has been repeated with controls, here PMID 11462117 Jagra (talk) 11:01, 11 December 2007 (UTC)

There was no placebo control for the dental amalgam removal. But the main problem is that metal sensitivity is not a recognised medical condition. It is fringe science, and would require a lot more evidence before putting it in the article. —Preceding unsigned comment added by Sciencewatcher (talk • contribs) 18:29, 11 December 2007 (UTC)

Sciencewatcher, I dont think your argument on metal sensitivity as fringe would hold up to any scrutiny. A pubmed search of metal sensitivity and dental will yield 336 RS papers and this is a ’lot more evidence’. Even if your argument were true this does not exclude the papers on amalgam from references in wiki Articles (see below for details). As only one word ‘amalgam’ is intended this is hardly undue weight, and remember it is under a heading. Not as proven not as probable (statistically speaking) but as POSSIBLE effect on the course of CFS.Jagra (talk) 01:37, 13 December 2007 (UTC)

Again Jagra you try to win the argument by weight of references. I suggest you read all 336 of those papers and see [1] if they are well-designed studies and [2] if they prove the conclusion you are suggesting. Also, try searching for homeopathy and see how many references you get. --Sciencewatcher (talk) 16:08, 14 December 2007 (UTC)

Well it is better than some who provide little in the way of references and mostly air opinions. I have never indicated any interest in homeopathy, it is you who keeps bringing it up, wonder why? It was your claim that metal sensitivity is fringe science, a statement you provided no evidence to support, when the weight of evidence as I have shown disputes that easy claim. It is for you to prove your claim not me, and so far you have not convinced me. Furthermore, the inclusion of metal sensitivity papers in journals such as; Evid Based Dent, J Periodontol, Dent Mater, J Prosthet Dent, Dermatol Ther, and J Oral Rehabil among others would confirm that it is accepted in the mainstream and therefore not the fringe science you claim. Just what placebo control do you think is appropriate for amalgam removal anyway?Jagra (talk) 11:39, 16 December 2007 (UTC)Jagra (talk) 04:02, 16 December 2007 (UTC)

Homeopathy is just a good example of something that is clearly quackery but has an awful lot of references. The placebo control would have to be a procedure that the patients thought was amalgam removal, but where nothing was actually removed. --Sciencewatcher (talk) 16:31, 16 December 2007 (UTC)

Clearly not practical as the amalgam would still need to be exposed to the mouth and therefore be visible. Unless you have an answer, it is an unreasonable expectation for placebo control and the studies have done all that can be expected. Furthermore metal sensitivity is not fringe science and the Melisa test has been validated PMID 12743534 and controls used in CFS studies. Jagra (talk) 11:00, 18 December 2007 (UTC)

I agree it isn't easy having a placebo control, but it is necessary in order to validate the results. --Sciencewatcher (talk) 15:45, 18 December 2007 (UTC)

This is one of the big methodological problems with the allegedly reliable and substantial results from the CBT and GET studies. What are the placebo controls for CBT and GET, and how are they double blinded? The answer is that there are none, and hence they can't be. Bricker (talk) 16:26, 19 December 2007 (UTC)

With CBT you are dealing with a psychological treatment, so essentially the treatment itself is a placebo effect. But I agree that it should be compared to other psychological treatments in order to find the one that works best. As people have pointed out, there are some major flaws and ambiguities in the theory behind CBT as it is applied to CFS, and GET has similar problems. But as for including them in the article, both CBT and GET are used as standard treatments in the UK and other countries, whereas other dubious treatments such as amalgam removal have no mainstream backing. --Sciencewatcher (talk) 00:04, 20 December 2007 (UTC)

No argument that CBT and GET should be included in the article (with appropriate caveats about methodological problems, limited results, patient dissatisfaction, and in the case of GET the potential dangers). But as I understand it (and leaving aside the previously discussed issue of placebo itself), the general debate about whether CBT is essentially the placebo effect is far from settled, though it certainly is an interesting possibility. Even if it is the placebo effect, it still cannot be properly double-blinded, which is an important component of control in clinical trials. That said, I don't think we can ignore any studies simply because they don't control for placebo effect, though it certainly reduces their rigour and validity. Bricker (talk) 07:31, 20 December 2007 (UTC)

I would agree that anywhere placebo control is impracticable that adequate caveats should definately be included and with that proviso, should not be excluded, Jagra (talk) 11:31, 20 December 2007 (UTC)

Unexplained and persistent fatigue

Without wanting to add to controversies, I think it might be wise to clarify at this time, for all concerned, whether or not this Article encompasses ’unexplained and persistent fatigue’. For those not familiar with the term and its historical synonyms, such as ‘CFS like’ or ‘resembling Chronic Fatigue Syndrome‘ this is code by researchers for “CFS that does not meet the 1994 CDC research criteria” But today may well meet the NICE guidelines for CFS diagnosis, and is not to be confused with ‘chronic fatigue’ per se. Serious CFS researchers (such as the CDC) are now carrying out parallel studies on this category and finding some distinct features (such as different gene expression), it may well end up being called CFS type 3 or the like. There is a body of evidence already out there and so it needs some consensus as to how it should be referenced in the Article, rather than ignored, as if it belonged in a different Article! My suggestion for what it is worth, is to distinguish such in the text (as we sometimes do for ME) For instance “this is found in CFS and/whilst in UPF that ---etc” Jagra (talk) 09:55, 6 December 2007 (UTC)

Jagra, there should be a "CFS vs fatigue" section that explains the differences. We have previously discussed the unreliable and arbitrary nature of CFS definitions, but discussion also faces the problem of interpreting the nature of the symptoms; physical and emotional symptoms can be misinterpreted on questionnaires, which is made more difficult when considering possible overlap and interaction. You may also be interesting in or already know that at the 8th International Association for CFS-ME conference, Dr Porter discussed the need to have different ways to describe "fatigue". Her study (with 200 controls and 120 with CFS) used factor analysis to differentiate various types of fatigue and found 5 terms based on subject reports: wired fatigue, brain fog fatigue, molasses fatigue (extreme heaviness), flu fatigue and post-exertional fatigue. Only flu fatigue was found in the normal controls. I think something similar exists for pain too, but "general aches and pains" is about as specific as health professionals usually get with CFS in practice. Perhaps more comparison needs to be done with fatigue and pain in CFS vs other illnesses. - Tekaphor (talk) 05:15, 7 December 2007 (UTC)

Tekaphor, interesting about different types of fatigue, and I agree with your suggestion for a 'CFS vs chronic fatigue' section and concerning practitioner diagnosis. But same cannot be said for experienced CFS researchers who are familiar with CFS fatigue, but if a patient has say only three not four additional symptoms from the list then it is not CDC research definition CFS but now is 'undefined and persistent fatigue' or 'CFS like' Could even meet NICE criteria, so also a double edged sword if labelled chronic fatigue, as no longer could it be claimed CFS is not chronic fatigue. Which side of the fence does it then fall, as far as the Article is concerned? Jagra (talk) 10:54, 7 December 2007 (UTC)

Keep in mind that England is only a tiny part of the world, and even there nobody has accepted the NICE criteria. Guido den Broeder (talk) 14:01, 7 December 2007 (UTC)

If the patient has chronic fatigue but doesn't meet the criteria for CFS then the illness is generally called "chronic fatigue" (or sometimes "idiopathic chronic fatigue"). This is already mentioned in the introduction and in "CDC 1994 criteria" section. —Preceding unsigned comment added by Sciencewatcher (talk • contribs) 16:52, 7 December 2007 (UTC)

The point being these are different to ’unexplained and persistent fatigue’. For those not familiar with the term and its historical synonyms, such as ‘CFS like’ or ‘resembling Chronic Fatigue Syndrome‘ this is code by researchers for “CFS that does not meet the 1994 CDC research criteria” But today may well meet the NICE (or other CFS) guidelines for CFS diagnosis, and is not to be confused with ‘chronic fatigue’ per se. Something the article does not yet distinguish. Jagra (talk) 03:57, 10 December 2007 (UTC)

To further clarify, there is Research definition CFS, Diagnostic criteria CFS and Chronic fatigue. Jagra (talk) 01:39, 13 December 2007 (UTC)

'Unexplained and persistent fatigue' relates to a much larger group of people than CFS, as well as a myriad of causes. For these reason already, it would IMHO be a bad idea to include such a body of evidence in this article. Guido den Broeder (talk) 11:47, 16 December 2007 (UTC)

Guido, how do you differentiate 'unexplained and persistant fatigue of 6 + months duration' from 'chronic fatigue'? Do you think encyclopedia readers can differentiate their diagnosis of CFS from "1994 CDC definition" CFS?Jagra (talk) 03:43, 21 December 2007 (UTC)

"Chronic fatigue" is (1) a symptom, which is usually explained, and (2) a control group in CFS research, where it means "no CFS but chronic fatigue of unspecified cause" but usually excludes chronic fatigue from unspecified known causes.

"Unexplained and persistant fatigue" is a subgroup of all people with chronic fatigue. It excludes chronic fatigue from diagnosed causes, but not from undiagnosed known causes. It should exclude people diagnosed with ME, but usually doesn't.

"Chronic Fatigue Syndrome" is a working diagnosis for scientific research, with a number of very different definitions that are all but two (Oxford, NICE) intended to restrict the target group to mainly patients with Myalgic Encephalomyelitis, but often fail to do so. Note that not all patients with ME are continuously fatigued.

Can readers differentiate CFS from Fukuda? No, nobody can, since Fukuda is one of the definitions of CFS.

Can readers differentiate ME from Fukuda? Perhaps, but not thanks to Wikipedia, because there is no article on ME. Guido den Broeder (talk) 09:04, 21 December 2007 (UTC)

Placebo, what placebo?

http://www.slate.com/id/2176465/

Those who freely invoke the notion of placebo to explain away results might want to take note of the following studies on placebo and related methodological issues:

PMID: 8839653

PMID: 11372012

PMID: 11816328 ("CONCLUSION: We found no reliable evidence that placebo treatments in general had effects of clinical importance.")

PMID: 12007544

PMID: 12535498

PMID: 15161896 ("CONCLUSIONS: The reporting of trial outcomes is not only frequently incomplete but also biased and inconsistent with protocols. Published articles, as well as reviews that incorporate them, may therefore be unreliable and overestimate the benefits of an intervention.")

PMID: 15257721 ("CONCLUSION: We found no evidence of a generally large effect of placebo interventions. A possible small effect on patient-reported continuous outcomes, especially pain, could not be clearly distinguished from bias.")

PMID: 15266510 ("CONCLUSIONS: There was no evidence that placebo interventions in general have clinically important effects. A possible small effect on continuous patient-reported outcomes, especially pain, could not be clearly distinguished from bias.")

PMID: 15817527 ("CONCLUSIONS: Most randomised clinical trials have unclear allocation concealment on the basis of the trial publication alone. Most of these trials also have unclear allocation concealment according to their protocol.")

PMID: 16117920 ("DISCUSSION: Outcomes in randomised trials are often reported selectively.")

PMID: 17060210 ("CONCLUSIONS: The blinding status of key trial persons was incompletely reported in most randomized clinical trials. Unreported blinding may be frequent, but one of five 'double blind' trials did not blind either patients, treatment providers or data collectors. Authors, referees, and journal editors could improve the completeness of reporting of blinding, eg, by adhering to the CONSORT statement. It is inappropriate to presume blinding of key trial persons based only on the ambiguous term 'double blind'.")

PMID: 17076559 ("CONCLUSIONS: This review classifies blinding methods and provides a detailed description of methods that could help trialists overcome some barriers to blinding in clinical trials and readers interpret the quality of pharmacologic trials.")

PMID: 17311468 ("CONCLUSIONS: This study classifies blinding methods and provides a detailed description of methods that could overcome some barriers of blinding in clinical trials assessing nonpharmacological treatment, and provides information for readers assessing the quality of results of such trials.")

PMID: 17440024 ("CONCLUSIONS: Blinding is rarely tested. Test methods vary, and the reporting of tests, and test results, is incomplete. There is a considerable methodological uncertainty how best to assess blinding, and an urgent need for improved methodology and improved reporting.")

PMID: 17517809 ("CONCLUSION: Two-thirds of conclusions in favour of one of the interventions were no longer supported if only trials with adequate allocation concealment were included.")

PMID: 17652297 ("CONCLUSIONS: The high proportion of meta-analyses based on SMDs that show errors indicates that although the statistical process is ostensibly simple, data extraction is particularly liable to errors that can negate or even reverse the findings of the study. This has implications for researchers and implies that all readers, including journal reviewers and policy makers, should approach such meta-analyses with caution.")

Bricker (talk) 23:42, 6 December 2007 (UTC)

There are just as many well-designed studies showing that the placebo effect does have a significant clinical effect over and above no treatment. Perhaps they just chose the studies that fit their pre-conceived conclusions? Or perhaps they chose studies where the placebo effect is likely to have little effect (such as cancer) instead of studies where it is likely to have a large effect (headache and depression). --Sciencewatcher (talk) 23:47, 6 December 2007 (UTC)

Some more thoughts on this: take the example of stress. Stress alters the immune system, causes headaches, increases blood pressure, and causes various other symptoms. It's pretty easy to see that a psychological factor (i.e. a placebo) that reduces the stress will also reduce all of these symptoms. So the question is: how could there possibly not be a placebo effect? --Sciencewatcher (talk) 23:58, 6 December 2007 (UTC)

I too used to believe the placebo effect was at least significant, and possibly powerful. But given the above evidence I have changed my mind. That's evidenced based reasoning for you. Did you even read those studies I quoted? A whole 5 minutes passed between my posting and your first response, and 16 minutes until your second response. No way you read through, understood, and carefully considered and responded to them in that very short time. That was a purely reflexive and ideologically driven response.

Other editors here will notice that SW's criticisms were not backed with even a single reference, a common tactic from SW.

For example, we are still waiting on the evidence for SW's claim about the "lots of autopsies" that show no brain damage; and some solid evidence about SW's overtraining claim, because the one reference SW gave us for that was laughable, and frankly more than a little hypocritical given SW's endless criticisms of his/her opponent's evidence, and demands they meet the highest scientific standard possible, and accusations of bias, etc. No, what we get from SW is just some slippery rhetoric about "Perhaps they just chose the studies that fit their pre-conceived conclusions?", and "how could there possibly not be a placebo effect?", and so on. Pathetic.

Bricker (talk) 01:14, 7 December 2007 (UTC)

Interesting article and studies Bricker! I also believed in the potential of the placebo effect, but it was actually CFS which challenged my pre-conceived notions that said illness would be resolved through stress management and cognitive changes and exercise. I am still somewhat reluctant to dismiss the placebo effect, but I get the impression that even when observed, it usually seems to have a temporary impact which eventually declines back to baseline. From many of the studies presented above, there seems to be a rather mild placebo effect on subjective measurements (especially pain), but these effects couldn't be clearly distinguished from bias, which itself is mentioned as a problem in so-called "randomised trials". As the Slate article implies, while there are subjective psychological effects in placebo studies, the objective improvements are lacking. We see something similar in CBT research for CFS; where some people say they feel subjectively better, but they lack the expected equivalent of objective improvements in performance. While I agree that Sciencewatcher's rapid response and accusation against the placebo researchers are suspicious, some of SW's other statements about the types of illness involved in the placebo studies warrant further consideration. - Tekaphor (talk) 04:56, 7 December 2007 (UTC)

Well Bricker, you are right, I did just spend 5 minutes reading the article regarding the 130 trials investigated. Perhaps that is because there is only the abstract to read! Assuming you have purchased the full article (which you would have to in order to see exactly what they were looking at), you would be so good as to give more details so we can discuss it properly.

It is very convenient that all the references you posted are by the same researcher! As I said, the placebo effect has very good science behind it, so you would need to have very good evidence to debunk it. Do you have that evidence? Or do you just have something against the placebo effect? --Sciencewatcher (talk) 15:54, 7 December 2007 (UTC)

You have missed the point, SW. It is not possible to determine the degree of placebo effect (if any) in trials that only have an active treatment, and a placebo arm. It is also necessary to have a no treatment/no placebo arm to compare against placebo, and when you look at all the studies that do have that form the result is no significant difference between placebo and no treatment. That is the evidence. Any other form of placebo controlled study doesn't count in determining the role of placebo. In light of the above work I cited, it is no good simply asserting placebo exists and there is lots of good science behind it, and accusing me of bias against the placebo (as I already stated, I used to believe it was real and significant). You have to show us the evidence. Or do you have something against the possibility that the placebo effect is not true, or at least far less significant than previously believed?

And we are still waiting for the evidence on autopsies, and overtraining, amongst other claims you have made. If you can't provide such evidence, then you should withdraw your claims.

Bricker (talk) 00:42, 8 December 2007 (UTC)

Bricker, as I already said above, there are lots of well designed studies that compare NO TREATMENT (just in case you miss it again) with placebo and show a significant effect. Did your researcher miss all of those? That is basic science, and I'm not going to waste my time pointing out the obvious, especially to someone who tells me to "sod off" and that my references are "laughable". I find it interesting that you told me to "sod off" when I simply mentioned the word psychosomatic in the same sentence as CFS! Clearly you have a problem, and no amount of evidence is going to sway you from your beliefs. --Sciencewatcher (talk) 16:23, 8 December 2007 (UTC)

"Clearly you have a problem, and not amount of evidence is going to sway you from your beliefs". Well, I guess I takes one to know one, sciencewatcher.

Additionally, we all have a problem with a repeat pattern in your rehtoric, namely making exaggerated claims ("lot's of autopsies", "CFS = Burnout") or airy claims ("NICE should say it's psychosomatic") and failing to come up with anything other than more rethoric when probed in depth.JayEffage (talk) 16:45, 8 December 2007 (UTC)

Exactly.

"Bricker, as I already said above, there are lots of well designed studies that compare NO TREATMENT (just in case you miss it again) with placebo and show a significant effect."

So where are they? Once again you make a claim and fail to back it. The studies I quoted about placebo did precisely what you claimed they didn't do, looked at ALL the eligible studies of that form, and found NO significant difference between placebo and no treatment/no placebo arms.

And we are still waiting on the evidence for the autopsies, over training/burnout, and also an explanation of how stress is a major precipitator of CFS without any objectively measurable changes to the stress system occurring until well after onset, amongst other problems with your various claims.

Bricker (talk) 23:27, 8 December 2007 (UTC)

Bricker, I could easily give references for all of those things, but what would be the point? You are openly hostile to me and have personally attacked me, and those things aren't related to proposed changes to the article anyway. I have better things to do with my time than to dig up references that you will probably ignore. For everyone else here, I suggest you look at the placebo effect article on wikipedia, and send me a message if you want any more info/references, etc., or if you want to discuss this further. The placebo effect is a fascinating subject, with too little research being done into how it works. --Sciencewatcher (talk) 19:52, 10 December 2007 (UTC)

-

HYPOCRITICISMS

I'm not "taking sides", I just have to point out what I see as hypocritical behaviour. Sciencewatcher, after that obvious knee-jerk reaction against Hrobjartsson's and Gotzsche's (etc) research, why should Bricker even bother following your suggestion? There's simply no way you read a 2 page news article and 16 abstracts and posted a fair response on the talk page all in 5 minutes. I also wonder if you have actually read the entire placebo page for yourself; some of the research you dispute is mentioned there, and as the article describes "These results suggest that the placebo effect is largely subjective. This would help explain why the placebo effect is easiest to demonstrate in conditions where subjective factors are very prominent or significant parts of the problem." Perhaps the overall study conclusions were too restrictive, but the rest of the page doesn't deviate radically from that statement. Their work isn't only about the placebo effect, but also looks into the common weaknesses in outcome measurements etc used in trials, but you should already know that since you read all those abstracts. They also point out that many studies have a small sample size and the placebo effect is weaker in larger trials, suggesting bias. Several of the abstracts have the full text freely available, while the full text of the main 2001 study [1] is available to registered users (free to register). It seems apparent that you instantly dismissed all of these studies as flawed without actually taking a look at them, probably because you didn't like the implications. This is something you have often accused others of in the past, but this time you have been caught out ...

You have also been hypocritical about the quality of your own references. For example, during the previous arguments and edit war on depression in CFS, the only reference you came up with was essentially someone’s website claiming that the prevalence was 65 to 90%, which was promptly reverted. I don't really care about the overtraining reference since you didn't try to place it in the article, but it was of a similar quality (someone's opinion or synthesis). At other times you have dismissed research presented by other editors on the grounds of being preliminary or inconsistent, yet relied on such research yourself when it suited you (usually relating to the biological effects of stress in CFS, but at least you have since conceded that the HPA axis findings are indeed inconsistent in burnout and CFS) ...

You displayed an impressive skepticism towards the unconvincing acupuncture study, but I wish you applied such stringent standards to your own sources. Personally I believe we need diversity on this page, so I actually appreciate your contributions to this article to some extent even if I don't always agree with them. I agree that the placebo effect is interesting and requires more research, but I also think that you have an exaggerated perception of it; similar to the oversimplified superpowers you attribute to stress and the fixation with the HPA axis as the gatekeeper to everything that could possibly go wrong in CFS. All experience and perception is likely to have physical correlates, so there is going to be grey area between "subjective" vs "objective" measurements of the placebo effect in disease. Additionally, some treatments are simply very difficult to adequately compare with an equivalent placebo, such as the benefits of graded exercise. The established existence of a mind-body connection doesn't automatically rubberstamp speculation about what the placebo effect can do to the body, there is obviously something to it but the actual details still have to be worked out ...

Tekaphor (talk) 03:06, 11 December 2007 (UTC)

-

Thanks Tekaphor, that's exactly what I was looking for. I assumed you need to subscribe to be a resistered user. I'll have a look at the article. Regarding your other points: I certainly expect people to post valid criticism to my edits, and remove dodgy references. But I won't respond to people like Bricker who tell me to "sod off" and that my references are laughable. --Sciencewatcher (talk) 18:05, 11 December 2007 (UTC)

This is not only about your edits but about your very own style of rethoric in which - and this is obvious to a number of people involved in this discussion - you apply double standards: One for everything that furthers your POV and another one for everything that doesn't. It can't go on like this.JayEffage (talk) 10:35, 12 December 2007 (UTC)

Having looked at the study, it doesn't really say anything new. It doesn't actually conclude that there is no placebo effect - it says there is a significant placebo effect for pain, but not for anything else. I would also say there is good evidence that there is a significant placebo effect for headache, depression, and psychosomatic illnesses. Have a look at this study which actually imaged the brains of patients and found differences in brain activation that corresponded with placebo response. This study showed that the placebo effect takes about 8 weeks to take effect for depression (the same amount of time it took the active treatment to take effect) as compared with an immediate effect for a pain placebo. --Sciencewatcher (talk) 21:25, 11 December 2007 (UTC)

I agree that the Hrobjartsson placebo studies aren't radical, although they are rather sobering. As for the depression study you mentioned, it is interesting that the placebo responders (38%) and medication responders (52%) had the same reduction in the Hamilton scale but demonstrated different changes in brain function. It would have been better if they also considered the following:

• measurements of participants' "belief" in the treatment (possibly boosting the placebo effect);
• measurements of self-sabotaging cognitions (possibly countering the placebo effect);
• extend the trial to week 12 after everyone was unblinded at week 8 (to see how the results are affected).

Depression can be quite reactive to environmental cues so it would also be interesting to know how much of that played a role, especially since the effectiveness of "happy pills" is promoted throughout our culture. Many CFS patients seem to accept a psychological explanation for their symptoms, and many believe in nutritional and alternative therapies, so I'm sure that most CFS patients have rather high expectations of the treatments or lifestyle changes they try, especially in the early stages of the illness. Unfortunately these expectations often don't pan out in reality, so some longer-term patients probably become cynical about any new treatments. However, I suspect that the lowered placebo effect in CFS is mostly due to the related biological dysfunctions rather than reduced expectations; in depression where motivation can be much more of problem than in CFS the placebo effect still seems to be going strong, while in CFS it is substantially reduced.

Tekaphor (talk) 13:31, 12 December 2007 (UTC)

I agree that more research should be done to see exactly what factors result in a high placebo effect. As far as I know, the Leuchter study is the only one that has actually imaged the brain of placebo responders. Knowing the psychological factors that cause the placebo effect would be useful in developing more effective psychological therapies for depression. A study by Khan (PMID 10768687) found that 75% of the effectiveness of antidepressants was due to placebo, although the placebo effect went away down when the trials were done a different way so it's hard to draw any conclusions (there's a free full-text of this study somewhere, but I can't locate it right now). --Sciencewatcher (talk) 17:13, 12 December 2007 (UTC)

Tekaphor, with psychiatric conditions, some effects that are attributed to placebo are no placebo effects at all, but represent a curative effect from attention, recognition, peace of mind, etc. This explains why some people were very surprised when it was found that placebo effects in CFS trials are usually low, lower even than with other physical diseases. Guido den Broeder (talk) 01:22, 16 December 2007 (UTC)

Guido: but that is the placebo effect. Why do you think there is a difference between what you have described and the placebo effect? --Sciencewatcher (talk) 16:29, 16 December 2007 (UTC)

Wilson's temperature syndrome

Please add some link to Wilson's syndrome (Wilson's Thyroid Syndrome or Wilson's Temperature Syndrome) from which I quote:

In March of 2006 an article for Wilson’s Temperature Syndrome was published on PubMed. Friedman M, Miranda-Massari JR, Gonzalez MJ (2006). "Supraphysiological cyclic dosing of sustained release T3 in order to reset low basal body temperature.". P R Health Sci J. 25 (1): 23-9. PMID 16883675. —Preceding unsigned comment added by Sohale (talk • contribs) 14:18, 8 December 2007 (UTC)

Antidepressants

There is a lot of unnecessary detail, duplication, poor referencing (no PMID numbers) and bad spelling/grammer in this section. I started tidying it up, but realised that it would probably be better just getting rid of a lot of it. This section really should just be a couple of paragraphs long, like the other medication sub-sections. --Sciencewatcher (talk) 20:48, 12 December 2007 (UTC)

I have now reduced this to two paragraphs with the basic info and removed all of the unnecessary stuff. If you have any objection, let me know. --Sciencewatcher (talk) 20:53, 12 December 2007 (UTC)

Chinese

Copied from above:

JFW: "I have my doubts about the verifiability of non-English references. Guido: "It goes against Wikipedia to assume that only English sources are verifiable.

JFW keeps deleting a section because it contains a source in Chinese. I have yet to hear a valid argument. Guido den Broeder (talk) 23:39, 12 December 2007 (UTC)

See the next section. "It goes against Wikipedia" requires urgent clarification. JFW | T@lk 00:57, 13 December 2007 (UTC)

Hardly. The main rule of Wikipedia is that statements need sources. The verifiablity directive explicitly states that such sources can be in a foreign language. It is not necessary that you, JFW can verify a source, let alone that it should be easy. Guido den Broeder (talk) 01:24, 13 December 2007 (UTC)

The concern here is that if there is no source that is sufficiently verifiable, the content is probably not suitable for inclusion. I'm patiently waiting for the translation of the articles in question. It is necessary that some people (hopefully including me) can verify the source. At the moment we are not meeting that criterion. JFW | T@lk 21:03, 13 December 2007 (UTC)

Chinese

Orangemarlin (talk · contribs) removed the acupuncture studies, because the only available source seems to be a set of two articles in a Chinese medical journal. Guido den Broeder (talk · contribs) reinserted them, even when I removed them again on two occasions.

I find this inexplicable. This is a significant article on a medical topic, and no other articles of the same importance use references in foreign languages. There is also absolutely no indication how many people actually use acupuncture for case definition-positive CFS/ME. JFW | T@lk 23:40, 12 December 2007 (UTC)

Interesting, but not an argument against this source. Guido den Broeder (talk) 23:45, 12 December 2007 (UTC)

Do we a) have anyone who can actually read the Chinese sources and b) any indication that the journals in question are actually reliable? JoshuaZ (talk) 00:48, 13 December 2007 (UTC)

Yes - send them on. But would it make any difference to your POV? Or merely offer another avenue to vent your POV if you decide to question reliability, statistical methods, parameters for choice etc etc? But by all means, copy the article here, and I shall have it translated it for you. docboat (talk) 03:56, 13 December 2007 (UTC)

The full article can be found here: [2]. Regards, Guido den Broeder (talk) 17:55, 13 December 2007 (UTC)

The source is a well-recognised university in Hong Kong - and the university is also a medical school. The journal is an organ of a society for Chinese medicine, and is supported by government in HK and Beijing. So the source is reputable. Working on the translation. docboat (talk) 05:10, 14 December 2007 (UTC)

Reliability is not even the issue. It is verifiability. Acupuncture use for CFS/ME could be widespread, but unless their popularity is documented, we cannot assess whether this is suitable for inclusion to begin with. There are 1000s of small studies on CFS, most of them methodologically weak. Why quote these two, in a language not accessible to most readers, on a treatment that doesn't seem to be popular, only because they have a PubMed abstract? JFW | T@lk 00:53, 13 December 2007 (UTC)

That's a good point. JoshuaZ (talk) 00:58, 13 December 2007 (UTC)

The sources are referenced to because they are methodologically sound and provide the best known evidence, which is more than can be said for many other proposed treatments. Popularity of all treatments for ME/CFS is poorly documented, as well as for most other diseases I might add. Chances are that acupuncture is by far the most popular treatment among CFS patients. China has a huge population, after all, and it's mostly in the right climate zone. Among patients in the west it is known that acupuncture treatment is available, and more than a few patients here try it or consider trying it. Guido den Broeder (talk) 01:17, 13 December 2007 (UTC)

My reverts were based upon both the language and on the reliability/verifiability. Not withstanding the personal attacks of GvB, my reverts were done according to every Wikipedia rule on NPOV. OrangeMarlin ^{Talk• Contributions} 01:27, 13 December 2007 (UTC)

But not on your personal knowledge of the information contained therein? docboat (talk) 03:56, 13 December 2007 (UTC)

It might help if you would provide us with your arguments so that an actual discussion can take place (that one little rule that you forgot about...). Guido den Broeder (talk) 01:34, 13 December 2007 (UTC)

Guido, I provided a number of reasons above why the studies are methodologically unsound, and you didn't have any answer. But you then re-revert the changes (as well as other changes I made).

Guido is being a disruptive influence here - falsely accusing me of vandalism, falsely accusing OrangeMarlin of not talking on the discussion page, and constantly putting up nonsensical arguments against removing an obviously flawed study.

As JFW says, there is no other evidence that acupuncture is in any way helpful for CFS, so you can't rely on a single flawed study. I'm happy for the article to say that people use alternative therapies such as acupuncture for CFS (and I think my edit already said that anyway). --Sciencewatcher (talk) 16:09, 13 December 2007 (UTC)

This user is on my ignore list since 20071213. Guido den Broeder (talk) 17:21, 13 December 2007 (UTC)

Ignore lists are rather silly, aren't they? All they prove is that one editor cannot face the possibility of having to address the other editor's comments.

I think my arguments stand. Guido is quite right that we have little proof of the popularity of all sorts of other treatments. I'm actually in favour of removing all sorts of other treatments that are based on one or two small studies that have methodological issues (e.g. the essential fatty acid stuff). It would have been rather nice if we had a secondary source that assesses the relative usefulness of all those treatments. Oh wait, that was the systematic review that formed the basis of the NICE guideline! Never mind, apparently that review and the NICE guideline are not used by anyone and cannot be reliable. JFW | T@lk 21:0

On Wiki3, 13 December 2007 (UTC) This user is on my ignorelist since 20071213. Guido den Broeder (talk) 21:29, 13 December 2007 (UTC)

This is all very entertaining, but not very useful. How about we just get a neutral admin to make a decision?

I agree with JFW's comment about treatments. There isn't any treatment at all for CFS that has good evidence for effectiveness in all patients. We should definitely remove the ones that only have small/poor studies behind them. --Sciencewatcher (talk) 21:42, 13 December 2007 (UTC)

Surely you do not have to resort to Chinese language articles to back up alternative medicine treatments for CFS. This is English Wikipedia, and as many of the articles as possible should be in English. This is true of all English Wikipedia articles, but strikes me as quite irresponsible in the case of medical articles, which whether people should use to inform them of medical options or not, we all know that they will use. I think that this is a bad idea.--Filll (talk) 23:48, 13 December 2007 (UTC)

The article contains a summary in English, with the relevant findings - standard practice for articles in a language other than English. But at request, I will read the article and pass on what information - if any - alters the meaning of the summary. docboat (talk) 05:13, 14 December 2007 (UTC)

Wikipedia is an encyclopedia, not a guideline; its task is to inform readers of knowledge, not of treatment options. The rule on sources in foreign languages is, that similar native language articles are prefered, but if these do not exist sources in foreign languages are acceptable. It is unfortunate that this particular article was only published in the Chinese version of the journal rather than in the English version, but that does not make it a bad source, just an inconvenient one. However, it may be that we find an equivalent article in English after all, or that one will be published later. The references are not set in stone, we can always change them if only for convenience's sake. Guido den Broeder (talk) 00:03, 14 December 2007 (UTC)

On Wikipedia, things are also done by consensus, and my reading of the situation is that you do not have consensus here. Sorry.--Filll (talk) 00:37, 14 December 2007 (UTC)

Concensus on rules is not made on the talk page of an article. It is irrelevant that some users here would rather have different rules. Guido den Broeder (talk) 00:41, 14 December 2007 (UTC)

Sorry I do not understand. Perhaps you are confused.--Filll (talk) 00:45, 14 December 2007 (UTC)

5 reverts within 24hrs is bad enough, but added is repeated disruption of talk page dialogue with "This user is on my ignore list...". Wikipedia is a collaborative process and failure to collaborate is therefore disruptive and incompatable with the project. User blocked David Ruben ^Talk 04:59, 14 December 2007 (UTC)

The message is in accordance with the shunning essay, which says that it is polite to let other users know if you are ignoring someone. Otherwise people would not understand why there is no response to that someone's talk. In short, its goal is to prevent disruption, rather than cause it. Guido den Broeder (talk) 19:45, 15 December 2007 (UTC)

Protection

Can we get the page unprotected now? --Sciencewatcher (talk) 16:22, 14 December 2007 (UTC)

Whilst I was not involve in that mop action, IMHO i'ld suggest not, but rather allow all editors to take the short breather to consider best phrasing, NPOV (ie ensuring non-trivial minority views not unfairly described), consider if any other reference sources that might be used etc. GdB block ceases tomorrow, and as a committed editor on the topic, probably should then allow opportunity for their (now) constructive engagement with other editors, and then if that happening and past edit warring finished then unblock (else not very stable to have an article dip back and forth into protection). David Ruben ^Talk 20:06, 14 December 2007 (UTC)

Okay, sounds like a good plan. Hopefully everyone here can make an effort to be more civil and reasonable in the future and avoid edit warring, me included. --Sciencewatcher (talk) 21:34, 14 December 2007 (UTC)

How about tagging arguably problematic chunks with the likes of {{Template:Disputed-section}} instead of ripping them out when there's no shortage of editors who will be very strongly inclined to immediately restore the arguably important content? -- Strangelv (talk) 06:23, 15 December 2007 (UTC)

I agree. That is indeed the correct way to handle this situation. Blanking, while other people just collaborately decided on a text, is not. Guido den Broeder (talk) 19:49, 15 December 2007 (UTC)

Of course, it requires that users that do dispute the section state their reasons, including why the replies so far did not change their mind, and attempt to convince the rest of us. At the moment, if I'm counting right, we have 5 or 6 users (WLU, Jagra, JayEffage, docboat, myself, and one unsigned) upporting the present text, and three different minority views of one user each in favour of blanking (JFW: no Chinese, OrangeMarlin: no fringe, Sciencewatcher: doubts statistics). So it's really up to these three users. Guido den Broeder (talk) 11:31, 16 December 2007 (UTC)

No. We have Guido, Jagra and (I think) Jay in favour of the current article, and me, JFW and Orangemarlin in favour of removing the bad references. I have given my reasons (above) about why the references are unsound and it is up to Guido or anyone else to discuss their objections if any. --Sciencewatcher (talk) 15:52, 17 December 2007 (UTC)

I don't support my own text, I just made it slightly more neutral than what was there before. I can't support a text that is generating so much dispute. Given the horribly wrong conclusions made by the authors of the Chinese studies, I think all that can be said is that acupuncture is being researched. Is this worth reporting when the conclusions drawn appear wildly out of keeping with the study results? Not sure, and why bother wasting so many electrons and bytes over an essentially worthless statement? I'm sure there are many treatments being researched for CFS, why should this one be pointed out? WLU (talk) 19:22, 18 December 2007 (UTC)

Interesting, to write something that you don't support. I'm curious now: what specifically, in your opinion, is wrong with the authors' conclusions? Guido den Broeder (talk) 19:39, 18 December 2007 (UTC)

<undent>I wrote the text based on direct reading of the sources. I've changed my mind given the outpouring of words from editors I trust and the contents of their arguments. But here's a list of problems I see (the first one turned up while searching for acupuncture)

1. Acupuncture has also been shown to relieve pain in fibromyalgia cases, and may be beneficial to CFS sufferers as well. from Chronic fatigue syndrome#Pain relief; it's unsourced, and it's a jump to go from fibromyalgia to CFS in my opinion. Are they the same thing? For most pages I'd tag that as WP:OR or WP:SYNTH, if not remove it outright, without citations to back it up or even prove that acupuncture can remove fibromyalgia pain.
2. PMID: 16248250 says CONCLUSION:EA on Back-shu points is one of the effective approaches for treatment of CFS. That's a very firm conclusion, particularly considering it's based on two checklists. Does it treat CFS, or the symptoms? There doesn't appear to be any control group, and this makes me wonder about regression toward the mean as well as placebo - symptoms come and go, and just having a doctor pay attention to a person can make themselves feel better (not aimed at CFS, this goes for all medical conditions as far as I know). As I said, a very strong conclusion for an uncontrolled single study based on a relatively small number of patients, and p<.01 makes the conclusion even more over-reaching for what its reporting. Also a rather poor sentence in english, making me wonder about the translation.
3. PMID: 17997936 says Conclusion:Acupuncture is a safe, effective treatment for CFS. How is that not over-reaching, for any single study (or in this case, two studies)? From my knowledge, treatment recommendations are made after programs of studies that last years, not two studies with less than two hundred participants. Further, this study confirms what I thought about regression to the mean - both groups improved, but one improved more. Hardly convincing, particularly when on one was p<0.05. How does this justify a conclusion of acupuncture being an effective treatment, let alone a safe one (where's the long-term follow-up and adverse effects? There's none in the abstract, what about dropout? What was their inclusion criteria and were individuals avoiding other types of treatment?) Finally, is skin penetration absolutely necessary for acupuncture to 'work'? From my reading, both groups got poked in the same spots, but one didn't penetrate the skin. What if just poking the same spot has the same effect as puncture? Why didn't their control group use totally different spots that were not in accordance with TCM theories? I thought this was a point of controversey regards the scientific examination of acupuncture, 'cause there's many ways to stimulate the acupressure points.

Methinks if someone submitted them to a journal with a higher impact factor, they would have very different conclusions, if it was published at all. Another reason I'm 'dis-endorsing' my opinion is I'm used to editing articles which are desperate for peer-reviewed sources, not stuffed to the gills with them. This is the first time someone has mentioned that pubmed alone might not be sufficient, and given the abstract I'm inclined to believe them. I'd always thought that criticizing sources was OR-ish, but apparently not. I'd love to see replication of this study by a different research group, that'd be more convincing. And has anyone posted this yet? Based on my reading of the table, improvements on physical fatigue were pretty close, 8.38 treatment versus 5.47 control (not to mention the control group started lower in the first place by almost 1.5 points), mental fatigue 7 treatment, 3.73 control, with control again starting over 2 points lower, concurrent symptoms 7.55 treatment, 7.56 almost identical, SF12 +6.93 treatment +3.73 control. Are the differences truly significant, fluke, faulty control, I don't know. Two studies indicates some research, but truly preliminary findings.

And that is why I have problems with the author's conclusions. They are far too bold, far too certain for what the studies are. WLU (talk) 20:18, 18 December 2007 (UTC)

I've never had any illusions about Pubmed; in my experience 75% of all articles there is worthless. This one (I'm focusing on the second article, which we've discussed most) IMHO belongs to the 25% though. About your concerns:

• From the results, acupuncture mainly treats the symptom fatigue. The authors do not claim that it cures CFS.
• Abstracts are often bold. Of course the authors know that this is just one study, but the journal wants to attract readers. Based on the outcomes of this study, however, the conclusions would be correct.
• The authors do not make a recommendation.
• That both groups improved is not very relevant. We see the same in many other treatment studies for CFS. A good percentages of patients improves over time. But in other treatment studies there is often no extra improvement from the treatment, while there is here.
• One variable is not significantly affected by the treatment. That does not make the study invalid. On the contrary, it seems to be in accordance with what we know about acupuncture in general.
• We quote many references that have no follow-up. Follow-ups are rare.
• AFAIK nobody dropped out.
• If 'just poking' would have the same effect than they would most likely not have found any significant results.
• The control group did not use different spots because that would be suboptimal with regard to checking for a placebo effect.

Guido den Broeder (talk) 09:22, 19 December 2007 (UTC)

• 'Regression towards the mean' is not an issue since (a) the two groups do not significantly differ from each other and (b) illness is not a random occurrence.

With regard to the first article, it is indeed important to know if there was a control group. If not, then that reference can be dropped. The last sentence at 'pain relief' is speculation and therefore needs to go. Guido den Broeder (talk) 12:14, 19 December 2007 (UTC)

Ok I'm going to jump right into the middle of this discussion. Regression to the mean is a random occurance and to some extent illness is too. Note I am not saying illness is completely random but through the course of any illness (especially chronic ones) you see fluctuations over time. People with chronic fatigue may experience good periods and bad periods completly independent of treatment. Furthmore you have to consider that you can never measure any illness reliably without error. This is especially the case for subjective tests such as "mental fatigue". I have never heard of a single psychological measure that has a reliability of 1. Given that you are not basing the results on the illness but rather a measurement of the illness, I think we can say that regression to the mean is most definately a factor. Does the study even report the reliability of the measures used? Do they reference past studies examining the validity of the measures used? All these factors are important in determining whether a study is valid. Information which the majority of the en.wikipedia community cannot get at because they do not speak Chinese.JamesStewart7 (talk) 11:25, 24 December 2007 (UTC)

The authors use the SF12 which is a validated questionnaire. Guido den Broeder (talk) 11:42, 24 December 2007 (UTC)

It's also a short form. As a general rule, short forms have lower reliability than long forms. The website on the form even alludes to this "Briefly, SF-12 reproduces the eight-scale profile with fewer levels than SF-36® scales and yields less precise scores, as would be expected for single-item and two-item scales. For large group studies, these differences are not as important, because confidence intervals for group averages in health scores are largely determined by sample size." Lower reliability suggests a greater role of regression to the mean. As there is more chance variability in measurement, a greater proporition of the variance can be said to be accounted for by factors that operate on purely statistical principles such as regression to the mean. Regression to the mean is definitely a factor in this study just like it is in virtually every other small to medium scale study. Even in large studies, regression to the mean occurs. It is just that in such large studies, with low confidence intervals, regression to the mean can be said to be accurately controlled for.

Now here are some other things I would like to know, but I cannot determine because I cannot read Chinese (btw I still want to know the Cronbach's alpha's for each submeasure; a competent reserach would report this information in the intro for past studies and the results for their current one). You can either answer this information, along with any future questions I and others may have, or admit that the study is not verifiable due to the language constraint and remove it from the article. All of these questions are relevant, since this is a primary source, so it is up to the wikipedia community to verify it. Were there any dropouts, if so how many? None are mentioned in the abstract which is somewhat atypical. It would be surprisign, bordering on suspicious if no one dropped out. Was an intention to treat analysis conducted to control for drop-outs? Did the study report the success of blinding (a big thing in an acupuncture study dure to the difficulty in making a satisfactory placebo? What were the statistical methods used to control the groups? Was it an ANOVA or paired t-tests or a non-parametric technique. If they used paired t-tests did they adjust the p value needed for statistical significance accordingly. Did they make any statistical adjustments for the fact that their groups significantly differed before treatment on all measures where statistical signifcance were found (if not I would throw this study out entirely as this is a huge oversight)? Did significance remain after these adjustments? What horrible randomisation procedures did they use that resulted in these unequal distributions? If parametric statistics were used, what were the skewness and kurtosis of both groups? The study was only single-blinded. What was the therapists allegiances to the therapy? Did the therapists have any investment in acpuncture eg run an acupuncture practice on which they are financially dependent? Did the therapists follow a set script to limit their verbal interactions with their patient or have someone else present (so as to prevent the therapist unintentionally giving away group assignment? That's all I can think of for now. Do any other scientists on wikipedia have some questions that they think are relevant to the interpretation of the results? Guido den Broeder, I hope I have addressed your concern that you have "yet to hear a valid argument" against the inclusion of the source on verifiability issues. There are obviously a lot of questions I need addressed to properly assess the scientific procedures used in the study.JamesStewart7 (talk) 02:55, 26 December 2007 (UTC)

Thanks, James. Since the two groups do not differ significantly, as explained earlier, there is nothing to research there. As I understand it, there were no drop-outs. I share some of your other concerns (and even the SF36 has limitations) but we would need to drop 99% of all references in this article if we were to apply such high standards and demand that everything in an article must be spelled out. Instead, we are simply careful in drawing conclusions, and e.g. don't claim that acupuncture has been proven effective. Guido den Broeder (talk) 11:56, 27 December 2007 (UTC)

Acupuncture _may_ be useful in treating pain in CFS, but that is as far as I would go. There is good evidence that acupuncture can treat pain (although fake acupuncture points also seem to have the same effect). But as far as curing CFS or even treating non-pain symptoms, there is no real evidence. --Sciencewatcher (talk) 16:20, 19 December 2007 (UTC)

More on acupuncture

I found a mention of a review protocol regarding acupuncture for CFS in the Cochrane Database: [3]. Is there someone who has access to the full text? Guido den Broeder (talk) 19:53, 15 December 2007 (UTC)

I don't have access to the full text either. However from reading the abstract it seems to be saying that they haven't actually done the review yet - I think the article might just be discussing the protocol they will be using, although I could be wrong.

Somehow I don't think it would be possible to have a review because if you search for cfs and acupuncture on pubmed you only get 7 results, and the only one with a placebo control is the one we have been hotly discussing here.

If you do want a reference, it might be better just taking one that doesn't have any placebo control (rather than a poor study that has one), such as PMID 17710799. But I only have access to the abstract of this article so I have absolutely no idea how rigorous the study is.

My inclination is not to have any references (as they all seem to be poor), and simply state that people use acupuncture, it benefits some, but there is no evidence either way about its effectiveness in treating CFS. --Sciencewatcher (talk) 00:00, 16 December 2007 (UTC)

Sciencewatcher, Although it does not specifically say acupuncture in the abstract, there is evidence that CFS patients frequently use alternative treatments and find them useful, and even preferentially use body-based treatments. PMID 17459162 (perhaps someone has access to the full text and can further advise.) Why stop at acupuncture when there is evidence that patients use anti-oxidants and find them useful, such as Ubiquinone (CoQ10) PMID 15889950 PMID 10882883 and a 90 patient placebo controlled randomised trial of Carnitine esters found benefit in CFS ? PMID 15039515 Jagra (talk) 04:31, 16 December 2007 (UTC)

Agreed. CFS patients use lots of alternative treatments and find them useful, and you can put that in the article. There just isn't any evidence of efficacy for any of them in treating CFS. --Sciencewatcher (talk) 16:36, 16 December 2007 (UTC)

Anyone? This protocol is likely to contain references. Guido den Broeder (talk) 10:11, 18 December 2007 (UTC)

Guido, Cochorane database protocol, the full text including references is here Jagra (talk) 11:53, 18 December 2007 (UTC)

But not for free, alas. I'm running on zero income. Guido den Broeder (talk) 13:48, 18 December 2007 (UTC)

Sciencewatcher, there is evidence for carnitine esters. Jagra (talk) 11:56, 18 December 2007 (UTC)

Carnitine is regular medicine, not alternative. Guido den Broeder (talk) 13:48, 18 December 2007 (UTC)

See my comment above in the anti-oxidants section. 6 patients in each arm of the study is nowhere near "evidence". --Sciencewatcher (talk) 15:47, 18 December 2007 (UTC)

Also, alternative medicine includes the use of treatments that are scientifically proven in specific contexts and for specific conditions, in totally separate contexts and for totally separate conditions. Niacin is an excellent treatment for niacin deficiency, but it's an alternative treatment for schizophrenia. One of the hallmarks of alternative medicine is prematurely transferring conclusions from petri dishes to people using anecdotal evidence. It's the use, not the substance, that makes it alternative. WLU (talk) 20:24, 18 December 2007 (UTC)

Carnitine deficiency as secundary problem to primary diseases is well-established. Guido den Broeder (talk) 20:42, 18 December 2007 (UTC)

The full paper shows two unpublished carnitine ester trials to establish the parameters (including placebo) for the published trial, on 90 patients, more than enough (see my response in that section). Carnitine is also considered an Anti-oxidant, If you prefere we can have a seperate anti-oxidant section, however in the 3 papers that will support patient use of alternative medicine, anti-oxiodants are included (as alternatives), including CoQ10 for which there is evidence also in CFS Jagra (talk) 11:06, 20 December 2007 (UTC)

See my comment above. You cannot "establish the parameters" with a small placebo trial. The 90-patient trial itself would have needed a placebo. Even the authors of the study say this in their conclusion! --Sciencewatcher (talk) 16:08, 20 December 2007 (UTC)

Then it is being studied.

WLU, fibromyalgia (FMS) is considered widely as often co morbid with CFS, and indeed the pain in CFS may be similar to some extent it can have considerable overlap. Acupuncture in FMS is more widely trialled and therefore those findings do have relevance. Indeed the benefits of acupuncture in FMS are not restricted to pain but also improve fatigue and anxiety. PMID 16770975 PMID 16541995 Similarily IBS (irritable bowel syndrome) is often co morbid with CFS and acupuncture has shown benefit here, distinct from placebo effect. PMID 18054727 PMID 17659687 It is not surprising therefore that acupuncture will have some benefits in CFS groups and as there is moderate evidence for acupuncture in chronic pain conditions more generally, IBS, and FMS, PMID 16734514 PMID 17713049 think it reasonable to say it is being researched.Jagra (talk) 02:56, 21 December 2007 (UTC)

Pain in Fybromyalgia is caused by a deviation in pain sensation. (Muscle) pain in ME/CFS is caused by lactic acid. If ME/CFS patients have comorbid Fibromyalgia, it stands to reason that they can profit from Fibromyalgia treatments of pain. IMHO it is believable that acupuncture can address pain sensation, and tiredness which is by definition a sensation, but not (so much) pain from lactic acid. Guido den Broeder (talk) 09:12, 21 December 2007 (UTC)

Hi Guido. Why do you think lactic acid causes the pain in ME/CFS rather than inflammation? I'm also interested in your thoughts about what causes post-exertional pain exacerbation in ME/CFS. A constantly elevated lactate level might not produce pain itself but could be indicative of other processes which cause pain. At least in exercise, lactate doesn't seem to cause pain and may actually be beneficial ( PMID 15308499 | PMID 16573355 | PMID 15131240 ). When patients talk of baseline muscle pain, are they speaking of the specific type of pain usually associated with delayed onset muscle soreness or the general type of pain usually associated with flu-like aching? I think this is an important distinction, although I doubt that post-exertion pain exacerbation in ME/CFS is exactly the same as the delayed onset muscle soreness which occurs in healthy people; if it was, that would suggest the dreaded "deconditioning" (although it could instead suggest a different mechanism which increases one's susceptibility to the effects of exertion). It has been my impression from the CFS article that post-exertional symptoms have so far been associated with oxidative stress, RNase L activity, and changes in gene expression. As for baseline pain, different types of pain seem to exist in different people with the same diagnosis regardless of whether it's "ME" or "CFS". Also, some type of "sensitization" seems to play a role in the baseline symptoms of both fibromyalgia and ME/CFS (even in Byron Hyde's literature), although perhaps in different ways, but I doubt this wholly answers the question of where the baseline pain is coming from. - Tekaphor (talk) 02:23, 22 December 2007 (UTC)

Nicely put Tekaphor as it shows the overlap in pain symptoms and sensitizations in FMS and CFS is not just a result of co morbidity. Which points to Acupuncture studies in FMS as relevant to CFS groups, co-morbid or otherwise. Regarding post-exertion symptom exacerbation in ME/CFS in this paper PMID 7968718 there is a description of exercise and post exercise changes in essential fatty acid metabolites, which shows an acute phase response with increases in LTB4 lasting several days post exercise . This is the same metabolite increased due to hyper-response (with associated oxidative stress) is vasaoconstrictive and also involved in inflammatory responses, (see here) which the paper suggests explains the increase. Not to forget the joint pain in CFS, more likely explained by inflammatory mediators/processes (I am aware of one study that found correlation of a higher incidence of disc degeneration in CFS) and also aware of CFS groups with this type of muscle trigger points not FMS, indicative of spasm (vasoconstriction!) generated and refered pain.Jagra (talk) 05:57, 22 December 2007 (UTC)

Muscle pain in ME/CFS is no different from the pain that healthy persons or athletes experience from exhaustion. It is part of the normal physiological process, where lactic acid builds up, except that in ME/CFS it comes much sooner due to the fact that the energy production is compromised. (There is other pain as well, indeed, from inflammation, infections and cardiovascular problems, but muscle pain is central.) Regards, Guido den Broeder (talk) 13:07, 22 December 2007 (UTC)

So you're saying that chronic muscle pain in CFS is essentially the same as the muscle burn felt during exercise and/or immediately afterwards? Interesting, it makes sense that people with CFS are in a constant baseline state of exhaustion, but I'm still unsure about the exact biological processes involved. I wonder if such a state explains why these researchers (PMID 15041034) commented that, "Most diseases are accompanied by a blunted response to acetylcholine but the opposite is true for CFS. Such sensitivity is normally associated with physical training so the finding in CFS is anomalous and may well be relevant to vascular symptoms that characterise many patients." However, some people may attribute this state to a "stress-induced muscle tension". - Tekaphor (talk) 01:11, 25 December 2007 (UTC)

Tekaphor, note the comment in the paper you cite of “separate function for acetylcholine and that is its role as an important and generalized vasodilator.” That ties in with the vasoconstriction I mention above. There is a connection but a bit involved for this discussion. Also training reduces the release of essential fatty acid metabolites from exercise. The fit are more resistant to developing post exercise symptoms than the sedentary, but athletes can be more susceptible to flu / infections. Training is the opposite to deconditioning, optimal level of activity in CFS is just below that threshold which will dynamically alter the disease state/ activity / recovery cycle. Note also the clue “higher fatigue and lower muscle pain thresholds to electrical stimulation” in CFS PMID 12531455 Jagra (talk) 08:24, 25 December 2007 (UTC)

Generally speaking increases in LTB4 are associated with reductions in PGE2 the first is vasoconstrictive the second vasodilative, so a double effect on constriction in some tissues and a singular effect in others. Acetyl choline induces increases in PGE2 and thus muscle dilation / relaxation. PMID 11023714 PMID 9227529 Post exercise a decrease in PGE2 is found lasting several days. Added to a lower baseline of acetylcholine in CFS means even more vasoconstriction and ‘muscle tension’ during recovery in CFS.Jagra (talk) 12:03, 25 December 2007 (UTC)

@Tekaphor: yes, that is correct. It is not true, however, that ME/CFS patients are in a constant state of exhaustion, except for very severe or badly managed cases. We do recover from exertion. The problem is that we (i.e. our ATP) get exhausted again very quickly, and recovery from overexertion (when ADP is used) takes time. It takes the body several days to restock on ADP. Guido den Broeder (talk) 10:21, 25 December 2007 (UTC)

Guido, Just to round off, it seems that ATP, ADP and PGE2 cause relaxation of smooth muscle at least PMID 6181532 so reductions in any of these may increase ‘muscle tension’ It also seems that ATP, and ADP stimulate prostaglandins (thus dilation and relaxation) and so PGE2 may be the effector for the ADP changes you mention PMID 92289 rather than or maybe in addition to lactic acid effects. Jagra (talk) 05:03, 26 December 2007 (UTC)

Stress and infection in CFS revisited

In continuation of the discussion in the "Overtraining and CFS" section of this talk page, I support Bricker's suggestion of editing the Onset section of the CFS article as a result of these present discussions. I've posted the related comments to a subsection of my CFS research userpage (Stress and infection in CFS) because it is rather large (8 paragraphs) and I didn't want to bloat up the talk page, so just treat it as if I posted it here. - Tekaphor (talk) 04:40, 22 December 2007 (UTC)

Wow. That is impressive work, Tekaphor, a major contribution to the discussion.

It is also some serious reading. I'll get back with some commments hopefully in 2-3 weeks (mid-January).

Thank you. Bricker (talk) 07:14, 22 December 2007 (UTC)

I second Brickers comments, very impressive, and found the discussion on viral / stress precipitation of CFS very informative. Second your proposed change Jagra (talk) 11:25, 22 December 2007 (UTC)

Agreed. At the moment the onset section is not very informative and can be misintepreted. Guido den Broeder (talk) 13:22, 22 December 2007 (UTC)

Yes, I agree as well. However you might want to remove the bit about starting in winter, as I think that might be redundant. Does research show that CFS is actually more likely to start in winter? --Sciencewatcher (talk) 15:21, 24 December 2007 (UTC)

Can we hold off making the change until about mid-late January? I will not have time to go over Tekaphor's comments in detail before then. Thanks. Bricker (talk) 16:00, 24 December 2007 (UTC)

Thanks! It surprised me how strongly specific infections stood out as a risk factor, I have learnt a lot about "ME/CFS" this year. Sciencewatcher, the references for the wording about winter are already in the article's Onset section. - Tekaphor (talk) 01:17, 25 December 2007 (UTC)

Thank you, Jagra

My warmest compliments to Jagra for finding suitable sources to support the use of "alternative medicine" by people with fatiguing illness. And it isn't in Chinese either! JFW | T@lk 08:04, 27 December 2007 (UTC)

Other medical treatments section

Overall I find this section way to optomistic about the efficacy of the treatments especially given this review: http://jama.ama-assn.org/cgi/content/abstract/286/11/1360 which concluded "OThere was insufficient evidence about effectiveness in the other 4 categories (pharmacological, supplements, complementary/alternative, and other interventions)". I highly recommmend that this review be included in the article. It is a secondary source published in a prominent medical journal and it analysed 350 trials.

First lets look at Essential fatty acids subsection. This section has presented the treatment as an open and shut case where essential fatty acid supplementation is clearly an effective treatment. I do not believe the reserach is nearly as clear as the section presents it and I think the wording should be changed to reflect this. In particular, I find this sentence troubling "This can be explained because these trials did not attempt to control for other intake or lifestyle factors that effect essential fatty acid utilization" Firstly I believe this interpretation qualifies as original research as I do not see a reference (the "this can be explained bit.." is OR, the rest may just be a fact.) Secondly, there are clearly other ways to explain the mixed results which have been left out of the article. Take for example this explanation "The second study was better controlled than the first and the observed significant effect of the first study was a result of poor controls". Of course this statement is OR too, but I'm sure the authors of the second study made comment on the differing results so their comments may be included instead. The point is the statement is purely speculative.

Then there is this: "Subsequent work by researchers at an Australian University of Newcastle has confirmed an alteration in the D6D enzyme activity". Confirmed? You mean found a statistically significant difference in D6D enzyme activity in one group of CFS patients. Scientists rarely ever use words like confirmed (btw how do I get a the original study, there is a conference preceeding reference only) and it seems like their only speculating an actual role of D6D enzyme activity because one study can hardly confirm anything. The sentence continues "and therefore an inability in CFS to produce sufficient immune metabolites, unless enzyme competition is reduced such as in the Martinovic trial using modulation." The reference is an article published in Med Hypotheses There's no published evidence for this (at least in this reference) and since the citation is 1995 there may even be evidence against it. Articles in Med Hypotheses are not suitable support for anything. They are hypotheses only.

Moving on. All the magnesium references are from the 60's. Why is this? There is far more recent research. The above noted review included some magnesium trials and the concluded there was insufficient evidence for its effectiveness.

Alternative medicine section. I removed the Chinese reference. I saw only one person on the talk page arguing to keep it and several arguing against. Such an unverifiable source is also a clear violation of Wikipedia:Verifiability so I consider it clear cut. I edited the section so that it states this "It has been found that CFS patients use alternative treatments and consider them useful" and I still find that sentence problematic (it was worse before). I have read many studies that surveyed patients that use complementary medicine. They all find that patients all consider these medicines useful regardless of their actual effectivness eg many patients think their homeopathy remedy is effective. This really needs to be pointed out or readers will conclude that these remedies are effective from this anecdote. I suggest that either this statment be placed into the wider context (everyone thinks everything they use is effective) and some mention be given to cognitive dissonance (if they use it the are more likely to think it works to justify it) or regression to the mean (patients get better naturally so they may assume it was the treatment or one of the million other cognitive biases that suggests these anecdotes are unreliable... or we can just leave out the statement altogether as it is misleading. JamesStewart7 (talk) 12:04, 27 December 2007 (UTC)

You can do pretty much as you want on Wikipedia, as long as you have more or more powerful co-editors than the opposition. That doesn't make it correct. Fact is, that many patients benefit from what you call alternative medicine, and many don't from what you call regular treatments, and that the definition of what is regular differs between locations and changes over time. Guido den Broeder (talk) 12:41, 27 December 2007 (UTC)

The Chinese references do not appear to be a violation of Wikipedia:Verifiability.

From WP:RSUE[4]:

"Because this is the English Wikipedia, for the convenience of our readers, English-language sources should be used in preference to foreign-language sources, assuming the availability of an English-language source of equal quality, so that readers can easily verify that the source material has been used correctly."

Therefore, without equivalent English language sources use the foreign-language sources.

"Where editors use their own English translation of a non-English source as a quote in an article, there should be clear citation of the foreign-language original, so that readers can check what the original source said and the accuracy of the translation."

Although this section talks about quotes, logically it should extend to any material from foreign-language sources. The references given were clear citations of the foreign-language original.

The assertion above, the review: "Interventions for the Treatment and Management of Chronic Fatigue Syndrome" analyzed 350 trials is misleading. The review paper identified 350 studies that could be analyzed, only 44 met the review criteria for data analysis and were analyzed. The smallest trial had only 12 participants which is a small sample to extract meaningful data from. Without the full paper to view it is unclear if, "There was insufficient evidence about effectiveness in the other 4 categories (pharmacological, supplements, complementary/alternative, and other interventions)" means there was not enough statistically significant results from the data to infer anything, or whether the results from the data did show any statistically significant lack of effectiveness for these 4 categories. Reading the abstract I would say the former, but I can not be certain. In addition, the study inclusion criteria is not available from the abstract nor is there any way to assess the study statement, "All conclusions about effectiveness should be considered together with the methodological inadequacies of the studies." The only thing this study seems to conclude is, "mixed results" and "Interventions which have shown promising results include cognitive behavioral therapy and graded exercise therapy." Ward20 (talk) 01:25, 28 December 2007 (UTC)

Ok how about we verify that the Chinese language is of equal quality to English papers before we include it. I want the questions I listed above answered so I can verify its scientific quality. Particularly what were their randomisation procedures and did they make any statistical adjustment for the fact that the groups significantly differed before treatment? If they didn't the study is worthless and it can be thrown out (and I suspect they didn't).

Onto the Jama article. 350 trials is how many trials were analyzed. It just so happens that most trials of alternative medicine are of poor methodolgy, hence whenever you do a review you have to throw out most of the studies. This btw, is the reason that I think this article should move to include more meta-analyses and less primary sources. We need someone to sift through the research and decide what studies are of poor quality (according to this review, most of them) and what studies are of good quality and thus warrant inclusion in the article. Since WP:OR states that we cannot do this, using a meta-analysis makes sense. Mixed results is also a perfectly appropriate conclusion and it is a far more sensible conclusion than what this article currently states. It turns out results aren't always clear cut. The current problem with the article is it pretends the results are clear cut when they are not eg fatty acids section. I would much rather the article say the results are mixed when they are then pretend that scientists have reached consensus when they have not. Also, please note that mixed reuslts is essentially the same as insufficient evidence when you're getting mixed results in small trials, as the paper suggested.

If 12 participants is too small a sample to get statistical significance with the expected effect size (see statistical power) I think all that shows is that the researchers are incompetent. The presence of bad, poorly constructed studies is not evidence for the effectivness of any treatment. Bad, poorly constructed studies that show positive results are also not reliable evidence for the efficacy of anything either. If nothing else, this study is evidence that many of the positive results for alternative medicine treatments in the area of CFS are unreliable and should be taken with the requisite amount of salt. I am instead looking at an article that references a hypothesis like it is the absolute truth.

Also something I missed before, Why is the article so skewed against CBT? The article takes a much harder approach to CBT which has at least some reliable evidence to support it than it does with essential fatty acid treatment which has comparatively weaker evidence. Take this for an example, "Cognitive behavioral therapy (CBT) is claimed to be an effective evidence-based therapy for CFS" Claimed? Shouldn't this sentence read "There is evidence to suggest that CBT is an effective evidence based therapy for CFS". Also the way the sentence is worded, it is like there is some doubt over whether CBT is evidence based. It is evidence based. The word claimed shouldn't even appear in a wikipedia article, except perhaps with things that have been thoroughly disproven. A lot of statements that cast doubt over CBT such as this one, "Some patients and advocates suggest that there are “good” and “bad” forms of CBT, and it is important for patients to decide whether CBT is advisable in their case" are lacking a reference. Note, I have no problem with well referenced and valid criticisms like this "A systematic review on CBT[178] finds that, while some kind of positive result is often reported, the quality of the research into the effects of CBT is usually rather low and the patient selection is not random". Can whoever wrote this "A large evaluation study in Belgium, however, lead to the conclusion that while on average CBT may cause patients to feel somewhat better, objective measurement shows no reduction in their disability" make some mention of what objective measurements were used. Also, what is "large"? (language issues again yet again).

I want this section gone "Cautions" or edited so that it refers to self-guided treatments only (as opposed to graded exercise therapy as administered by a licensed psychologist/doctor). It's unreferenced and POV and it may deter people from seeking what is an apparently effective treatment, "Gibson Report also states that GET is one of the most common treatments for CFS and found 50-70% of patients improved somewhat with GET". I'm starting to see a reason for all of this though. Not the POV in this sentence, "However, the report itself has been criticised by several groups, for: being poorly conducted, misrepresentations, omissions, lack of references, factual inaccuracies or bias, and even potentially damaging implications" The ME association seem to be responsible for the bulk of these criticisms and I can not see any specific scientific complaints with the articles (although I do see many ad hominen attacks). As the quote states that "Several groups"(weasel words btw) have stated the article is "poorly conducted", it would seem like some researchers has leveled scientific criticisms at the article. I can see not such complaints in the references and all the references are from special interest groups.

Overall I think the treatment section is written too much from the POV of the ME association and related groups, whom I understand are resistant to the idea that psychological treatments may be used for CFS as they define it as a biological but not neurological condition. They are also rather accepting of complementary treatments for the disease. Because of this I am tagging the treatment section for the inevitable POV dispute that will follow my comments here. I am also going to point out that WP:NPOV policy suggests that the opinions of all major parties be presented in the form of "So and so states..." It also suggests that scientific research is a relatively neutral source "Disagreements over whether something is approached neutrally can usually be avoided through the practice of good and unbiased research, based upon the best and most reputable authoritative sources available." thus reports on the findings of scientific research should not really be followed by non-scientific criticisms of that report from obviously POV groups.JamesStewart7 (talk) 03:09, 28 December 2007 (UTC)

I can't understand why a company that has an NDA with the FDA for CFS after over 10 years of testing isn't mentioned anywhere.

Here's one article: Hemispherx Biopharma Reaches Agreement With FDA on Specific Steps to Achieve Complete NDA on Proposed CFS Treatment http://biz.yahoo.com/pz/080306/137712.html —Preceding unsigned comment added by 75.84.156.180 (talk) 22:51, 12 March 2008 (UTC)

POV changes

Firstly I removed this "political inquiry into the science was a statement about the public meeting" due to innaccuracy. This was never a statment about the Gibson report as the article suggested. It was a statment about the public meeting.

I inserted this "The ME Assocation states that the report misportrays several aspects of chronic fatigue syndrome." Personally I do not even think that should be there, at least not in this section (elsewhere may be ok). It is a political statment about a science report. Previously it was phrased as a scientific criticism. It is, however, the opinion of non-scientists and I cannot find any section in which they point to scientific errors with the article.

I removed this "The discrepancy between trial results and patient group surveys has been noted by the P.A.C.E. trial group, who are conducting a larger more detailed study into CBT and GET which is currently underway and is due for completion in 2009." as it is misleading. Firstly it is unpublished research. They may very well find that th etrial results are accurate and the surveys are wrong. The trials are regraded as scientifically more reliable due to things like push polling and unrepresentative selection. The statement, however, suggests the survey is more reliable as the statment is used in support of this claim "The 25% ME Group Submission to Gibson</ref> they state that both CBT and GET are not only just unhelpful to many severely affected CFS patients but also dangerous/harmful". I think this whole paragraph still needs many more changes. There are a lot of people bashing the Gibson report without giving scientific criticisms yet there are no words from supporters or even any rebuttal. Some note really needs to be made of the fact that the ME group did not publish its dangerous/harmful claim in a science journal. In fact, it is not clear what scientific evidence this claim is based on, if any.

I removed a few long standing uncited and POV from the section about CBT.

"Those who fit a 2003 ME/ICD-CFS definition with post exertional malaise may wish to consider whether graded exercise is recommended in their case because it can cause serious deterioration in the exertional intolerant, and the 25% ME Group^[1] point out that many severe cases were in fact mild cases before undergoing such therapy." This is an opinion presented as fact. The opinion may have a place in the article (somewhere away from the evidence about effectiveness) but a statement like this needs to be supported with a scientific reference.

"However this level of efficacy was only found in several small trials and were not even compared with specialist medical care or pacing." I'm sure they had some form of control group. This statment is clearly POV ("not even"). If someone wants to reinclude a similar statement please be more specific and state precisely what the control group was instead of what it is not. If there is actually no control group (even no treatment control) the stating that will suffice.

"Its role in helping severely disabled patients has not yet been properly established, but uncontrolled studies suggest it can help so long as it is tailored to the individual patient." This is OR, and somewhat redundant. There are many claims above from referenced sources stating that the trials may not involve the most severely disabled. It should also be noted that this is the case for basically everything eg you don't do a depression trial on the patients that are a high suicide risk for ethical reasons. I think it is misleading to state that the worst CFS patients are not included in trials but fail to state that this is the case for practically everything.

"However the Gibson Report also mentions the 25% ME Group findings that only 5% of their members found GET helpful and 95% found it unhelpful; and while the report used the word "unhelpful," the 25% ME Group insists that GET can also be dangerous/harmful." "Findings" is an inappropriate term. It is a survey response from a group that is specifically made for people who are severely disabled and have presumably not benefitted from treatment (if they had benefitted they wouldn't be severely disabled). That's like doing a survey of religious observance in a church. Edited to reflect this. I also removed this "the 25% ME Group insists that GET can also be dangerous/harmful". Note there is still significant mention of possible risks in that paragraph. However, they are now "possible risks" as there is yet to be a scientific report confirming them.

Cautions section is gone. It is repeated so many times above and this section is unreferenced. Btw I think the whole section on CBT/exercise needs a cleanup as there is a lot of redundant info. This may go a long way to solving the article is too long problem.

""Pacing" is being advocated by many patients as one of the few really effective means of minimising homeostatic disequilibrium." Firstly "homeostatic disequilibrium" is an extremely vague term, so vague that it doesn't really mean anything. Secondly this quote is an opinion presented as fact. I would change it to the form of "patients state that such and such is effective" but I think the scientific evidence covers the effectiveness of the treatments much better than opinions do.

"A few patients believe they find health benefits and pain relief from gentle stretching, non-aerobic exercise, and gentle activity." It is now "believe they find". See cognitive dissonance, illusionary correlation, regression to the mean and list of cognitive biases if you do not understand why I made this change.

"It must be pointed out that some antidepressants can exacerbate symptoms, especially in the first few weeks of starting a new drug, and can induce muscle weakness, sleep-waking dysfunction and cardiac arrythmias, amongst other negative side effects ^{[citation needed]}" Removed. I would have just fact tagged it but different anti-depressants will produce different side effects and I do not think it is fair to group all anti-depressants together. If someone wants to do a properly refrenced run down of side effects per anti-depressant then fine.

"Some studies have shown a reduction in symptoms with MAOI's, Tricyclics, SSRI and SNRI use" this is well referenced. "Some studies have shown no improvement" This has no references and has been fact tagged. I do not want to leave that in there long without a reference. It shouldn't have been in there in the first place without a reference to support it.

OR section pain, sleep, antibiotics. The antibiotics section is especially troublesome "In the MedLine database, ciprofloxacine, doxycycline and the penicillins are reported to be of significant (albeit temporary) effect in some patients. An even larger group of patients may have adverse effects, and a third group no effect at all." This should definitely have a reference. Also I have to question the public health implications of prescribing anti-biotics without evidence of efficacy so this is perhaps something a wikipedia article should not advocate without scientific support.

"This can be explained because these trials did not attempt to control for other intake or lifestyle factors that effect essential fatty acid utilization." Before I stated I was going to edit this. Now that I have re-read it I am removing it completely. Previously I thought the first trial did the controls while the second did not but apparently no trial controlled for this. Random assignment should really even out other all other factors unless there is a large interaction (something of the order of there being no effect unless a particular lifestlye factor is present).

This has been removed "In the early 1990’s Martinovic carried out a clinical trial on CFS patients in which such factors were controlled for and obtained remarkable results, many of whom became fit for work, after about 4 months, and were still well 16 months later. The trial used essential fatty acid modulation rather than supplementation, in which the dietary intake was adjusted for each patient, and varied according to their symptom fluctuations. Exercise and activity, physical and mental was set according to patient abilities, such that exacerbation of symptoms did not occur and levels increased over time to set formula. Cognitive behaviour (CB) was used to show patients (make cognitive) how symptoms varied with activity and dietary changes. Also behaviour therapy, (BT) was applied so that patients learnt to adjust the factors for themselves. ^[2]" I find all of this highly doubtful and the only reference is a med hypotheses and not the original study. Also there is this "The protocol for this therapy has yet to be published." They couldn't put out a protocol in 15 years or so? I also did a search through medline for the original reference. It looks like all there is is a few case studies that were never verified. It is not unreasonable to base a hypothesis off a case study but the article talks like it is a large trial. I also think case studies have minimal relevance to a wikipedia article in which far more reliable evidence is reported.

"and therefore an inability in CFS to produce sufficient immune metabolites, unless enzyme competition is reduced such as in the Martinovic trial using modulation.^[2]" Removed. Hypothesis presented as verified reserach.

Lifestyle changes section has been fact tagged for more than long enough to find a reference. As it is unrefernced and perhaps damaging (it basically advocates giving up, suggesting that will help "Importantly, acceptance rather than "fighting" to be as healthy as the patient was before CFS onset will lead to less frustration and fewer relapses.") I have removed it.JamesStewart7 (talk) 04:32, 28 December 2007 (UTC)

Oh brother, here we go again.

James, you appear to have missed (or misread) the warning tag at the very top of the article about it being classed as controversial, and all that this implies, in particular the requirement to DISCUSS changes on the talk page BEFORE unilaterally making them in the article. Whatever fair points you might have made have been lost in your POV rant.

Unilateral, high-handed bulldozing and posturing as a neutral hard-nosed objectivist ain't gonna get you very far. I for one, don't buy it. You are not the final arbiter of what is scientifically acceptable and what isn't. I strongly suggest that you are not really properly familiar with the literature in this field, for example your claim about there being no serious expert criticism of CBT and GET. Try reading the Canadian Guidelines just for starters, they have some seriously critical comments about the efficacy and role of those therapies, and you cannot dismiss those authors as ill-informed POV warriors. Furthermore, the methodologies of your precious CBT and GET studies are nowhere near as robust as you would like to believe, just look at both the patient selection and the outcomes assessment criteria, absolutely saturated with subjectivity.

I find your last paragraph particularly ill-informed and offensive.

I vote we revert the changes James has made, and get those sections locked down, until (yet again) the relevant issues are thrashed out on this talk page.

What say the rest of you? Bricker (talk) 05:06, 28 December 2007 (UTC)

I agree that James appears to have introduced some pov of his own by making these changes. However, I believe that he has made an honest attempt to wikify the text and rather than reverting or critisizing the changes, I intend to go over the new version in its entirety and rephrase where necessary, while trying not to exceed the present length. Guido den Broeder (talk) 08:43, 28 December 2007 (UTC)

I would rather revert, discuss, then edit, but do not want an edit war. The large scale changes that were made without consensus were not extremely constructive so soon after protection expired. Nor do I consider many of the changes NPOV. The very first edit replaced wording very faithful to the source with weasel words. But, I do agree there are weasel words and words to avoid that should be changed. Statements that can not be verified should be removed and more statements need to be attributed to the source, "So and so states..." Ward20 (talk) 09:13, 28 December 2007 (UTC)

Not only did he alter the the conclusion of the cite to weasel words but also altered the words agreed in consensus on this page to suit his own POV. Not an auspisious start James I also note you base your conclusions on wording of the abstracts without actually reading the papers. Yes you may have to visit a medical library before you can reach conclusions like you have in places. This is a LONG Article and it would help if you actually provide headings for PROPOSED changes, as others have pointed out to you already it is a requirement here for discussion- consensus- change. I too will be reverting some of those changes where i think POV appears until you discuss.Jagra (talk) 09:58, 28 December 2007 (UTC)

Regarding your apparent unqualified support for CBT and GET, if you read this talk page i think you will find a recent consensus that as these clinical trials lack Placebo controls, they should only be included heavily qualified. There was to be a reveiw of that material intended anyway in the near future and it would have been more constructive for you to have participated in that or to have made your veiws known during the recent discussion. Clearly as you have now removed what ever qualifications there might have been, it might be easier to have removed the lot, as these studies are flawed for the above reason, not pointing this out demonstrates POV. Jagra (talk) 11:11, 28 December 2007 (UTC).

Look if anyone wishes to revert particular changes, by all means do so. I will not rerevert without discussion here. In future I will present proposed changes here first also. Ok onto the comments. Firstly I can see Ward20's case for the wording of http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=pubmed&dopt=AbstractPlus&list_uids=10882883 as it does follow the article quite closely. In retrospect, I should have discussed this change first. However, I wouldn't call this the most reliable source. THe source was published in integrative medicine, a journal which survived only three years and published articles such as" The Plausibility of Homeopathy The Systemic Memory Mechanism" in which it was stated that homeopathy is highly plausible (homeopathy however, provides the complete oppostive view and is well referenced). Due to the presence of articles such as this in the journal, I consider the source itself, POV. Other sources, such as http://jama.ama-assn.org/cgi/content/abstract/279/19/1548, give a different take and largely avoid any suggestion that the treatments are effective.

Re: CBT. It is common for pyshological trials to include non-treatment controls instead of placebo controls. This is largely due to an extreme difficulty in creating a placebo. Some trials, however, will compare other psychological treatments. Here were the previously included qualifications. There was this one that has been unsourced since April "Some patients and advocates suggest that there are “good” and “bad” forms of CBT, and it is important for patients to decide whether CBT is advisable in their case". There are all these ones which are scientific sources and are still in there "A systematic review on CBT[178] finds that, while some kind of positive result is often reported, the quality of the research into the effects of CBT is usually rather low and the patient selection is not random. The reviewers note that reasons for withdrawals typically remain unreported, and furthermore state that a degree of publication bias seems to be present. In one study, the effect of CBT has been demonstrated up to five years after therapy[191]. A large evaluation study in Belgium, however, lead to the conclusion that while on average CBT may cause patients to feel somewhat better, objective measurement shows no reduction in their disability", "A systematic review on CBT[178] finds that, while some kind of positive result is often reported, the quality of the research into the effects of CBT is usually rather low and the patient selection is not random. The reviewers note that reasons for withdrawals typically remain unreported, and furthermore state that a degree of publication bias seems to be present." This, however, is far too vague "However, the report itself has been criticised by several groups, for: being poorly conducted, misrepresentations, omissions, lack of references, factual inaccuracies or bias, and even potentially damaging implications" Poorly conducted how? Misrepresented things how? It is also from a non-scientific source but it seems like it is a scientific source because it discusses trial quality.

I do not wish to remove any valid scientific criticism for any treatment. I also do not show unqualified support for me. Looking at the evidence, I would actually say that it is unclear whether CBT is an effective treatment although there may be some (weak) support for this claim. I wholeheartedly support any move to include relevant scientific criticism of CBT, GET or any other treatment. Similarly any reviews finding negative results for CBT/GET should most definitely be included. I just don't think non-scientific and poorly specified criticisms of the treatment from non-scientific sources should be included. Such non-specific criticisms can never be countered. If people want to make well specified criticism about the CBT, GET studies such as some note of what the patient selection criteria involved as Bricker mentioned then by all means do so. Btw Bricker wouldn't the Candian guildine papers be much better references for the crticisms of GET?

I do, however, still stand by my comments about alternative treatments needing to be supported by evidence before any claims of efficacy appear in the article and I believe that moving long standing unreferenced claims is supported by WP:Reliable sources. Also while I may have violated wiki-ettique by not discussing these changes first, I would like to point out the controversial tag states " Make sure you supply full citations when adding information and consider tagging or removing uncited/unciteable information in highly controversial articles." and the majority of what was removed was uncited or was divorced from what the references say eg something that is a hypothesis in the reference becoming a fact in the article.JamesStewart7 (talk) 13:27, 28 December 2007 (UTC)

I just want to clarify one of my past points that I think may have been misinterpreted "Cognitive behavioral therapy (CBT) is claimed to be an effective evidence-based therapy for CFS"" I wouldn't have objected as strongly to "Cognitive behavioral therapy (CBT) is claimed to be an effective for CFS" as whether CBT is effective or not may be debateable (althought it may be stated better - claimed does appear on WP:Words to avoid for a reason). Problematically, however, that sentence implied two things; that it was only a claim that CBT is evidence based and it is only a claim that CBT is effective. It was the claimed it is evidence based bit I objected to. Also I did not mean to imply that as Bricker suggested I think there is "no serious expert criticism of CBT and GET" What I meant was that the criticisms of CBT which I removed were not expert criticisms. That is not to say that there are not expert criticisms (some of them are still in the article), it's just that there were a lot of non-expert criticisms that were not labelled as such. I hope that clarifies things. JamesStewart7 (talk) 14:23, 28 December 2007 (UTC)

Regarding PMID 10882883 I suggest you should also have looked into the credentials of the authors involved, two i am familiar with is J Goldberg and D Buchwald, both long associated with CFS research with 26 and 78 pubmed papers largely CFS related, to their names in a whole range of mainstream journals. In addition the twin studies is one of a series of articles of CFS aspects in studies involving these authors published in; Arch Intern Med, Sleep =2, J Psychosom Med, J Infect Dis, Pain Med. Regardless of your POV the journal is RS and V. Feel free to bring forth expert criticisms of CBT and GET.

I did not say the sources were not RS and V, I said they were POV. What is this idea that you cannot be a good researcher and have a POV. I am not aware of a person on the entire planet who does not have a POV. As I stated above, other sources who found similar rates of alternative medicine use (but not specific to chronic fatigue) did not use the pharse "found helpful" instead using terms such as "perceived benefits". How do you explain the differences in wording if it is not due to differences in the authors POVs? Note it is not their results that differ. What differs is they words they used when they reported their findings. Jagra as others have already built off this revision and reverted much of what they object to and since Sciencewatcher recently made a large edit I do not wish to revert. I'll leave it up to those people to do that if they deem it necessary.JamesStewart7 (talk) 01:52, 29 December 2007 (UTC)

I removed this reference before: "Gray JB, Martinovic AM (1994). "Eicosanoids and essential fatty acid modulation in chronic disease and the chronic fatigue syndrome". Med. Hypotheses 43 (1): 31-42" and I would like to point out this quote "One particular combination of interventions was used in a multi-agent adjuvant regime for CFS (unpublished work)... In summary, 27 out of 29 compliant CFS patients showed significant improvement in 3 months or less, when only 2 of those 29 had any improvement over the previous 12 months." This information was used to support this "In the early 1990’s Martinovic carried out a clinical trial on CFS patients in which such factors were controlled for and obtained remarkable results, many of whom became fit for work, after about 4 months, and were still well 16 months later" and was provided as indirect support for many of the other claims in that paragraph. Since the work was unpublished (and as far as I am aware still is) this seems to violate Wikipedia:Verifiability as no one can ever get to the original paper or sufficiently detailed information on how the treatment was performed.JamesStewart7 (talk) 02:18, 29 December 2007 (UTC)

If you want us to assume good faith you could start by reversing your reverts so discussion can commence. Jagra (talk) 23:51, 28 December 2007 (UTC)

I agree completely with that comment. James, the default position should not be that your edits stand until there is consensus on what changes are acceptable. I think you should revert your edits (why should anybody else have to do that work?), and then your proposed edits can be discussed and when (and if) a broad consensus is reached, then they can be put back up. You will certainly get a lot better reception if you did it that way.

It is also unacceptable to make an edit (especially without prior consultation and discussion) and expect others to have to rebut it straight away, and to then take a lack of immediate rebuttal as confirmation of the legitimacy of your edit. It takes time, energy, and considerable background knowledge to construct a proper rebuttal. (This is a common problem across Wikipedia, not just on this article.)

I notice that you are new to this article, James. This is one one the most controversial and poorly understood areas of medicine. The database is amongst the most confounded and patchy in medical science. There is serious disagreement between both patients and clinicians/researchers, and amongst clinicians/researchers (including over the nature and role of CBT anf GET). If you do not take the time to inform yourself properly about these very real problems, and take account of them, then you will not get far here and will just waste people's time, including your own.

That said, I agree that this article has many weaknesses, but maybe that just reflects the situation in the real world. Bricker (talk) 03:27, 29 December 2007 (UTC)

Ok I'm going to offer an olive branch and revert everything but the removal of long standing uncited information (most of it has been done anyway). However, I suggest people re-read over what the reference says here "political inquiry into the science was a statement about the public meeting" because I think it has been misquoted and I would like the essential fattty acid information looked at asap as the work is unpublished. JamesStewart7 (talk) 05:41, 29 December 2007 (UTC)

Weasel words again James this is what you just said “Ok I'm going to offer an olive branch and revert everything but the removal of long standing uncited information (most of it has been done anyway).” A reasonable approach but that is not what you have done is it? What should we make of that now! Jagra (talk) 08:30, 29 December 2007 (UTC)

I don't know Jagra? Maybe you could WP:AGF and assume that I missed something, as the compare revisions page isn't always easy to read and I stated I was trying to reinclude all the referenced material while avoiding the long tagged unreferenced material. Maybe you could also try and take a more constructive approach and point out what I missed so that I may correct it? Also btw Jagra it is not actually assuming good faith if I have to prove myself before you will do it. Really Jagra, you don't give me much chance to improve the article when you won't tell me what sections need to be amended. JamesStewart7 (talk) 08:48, 29 December 2007 (UTC)

Thank you for that.

I propose "Cognitive behavioral therapy (CBT) is claimed" be replaced with "Results from clinical studies has have suggested that Cognitive behavioral therapy (CBT) is", similar wording that you had, but slightly tweaked to indicate where the information originates.

About the argument, "As I stated above, other sources who found similar rates of alternative medicine use (but not specific to chronic fatigue) did not use the phrase "found helpful" instead using terms such as "perceived benefits". The two sources are not equivalent because one does not talk about CFS. When trying to introduce arguments from WP:RS that do not directly talk about chronic fatigue it usually becomes WP:SYN and WP:OR. I think this is very similar to a lot of the material removed in the antidepressants section.

However, using information from the CFS source I believe this is a good compromise, attributing the information with something like, In one CFS study designed to assess alternative medicine use and perceived benefit, 91% of the 63 study participants with CFS used alternative treatments, and a large proportion considered them helpful. Ward20 (talk) 08:00, 29 December 2007 (UTC)

Ward I have no objections to your wording for CBT studies. In regards to alternative medicine I was not trying to argue that my source be included but simply demonstrating that researchers may frame their findings differently. I would also agree that your proposed phrasing for the alternative medicince usage findings is a reasonable compromise.JamesStewart7 (talk) 08:48, 29 December 2007 (UTC)

Ward20 Your comments are well made and the suggested change to the first sentence wording of the Alternative medicine item is acceptable to me, with the second existing sentence added Jagra (talk) 08:30, 29 December 2007 (UTC)

Help requested on CBT ref ^[187]. It needs a conventional citation. The link provided will not show a reader without subscription any information about the reference. Ward20 (talk) 19:19, 29 December 2007 (UTC)

My suggestion would be to put "Conclusions from clinical studies" instead of "Results from clinical studies", since in many cases the results are not so clear as the authors think they are. Guido den Broeder (talk) 19:23, 29 December 2007 (UTC)

I understand your argument, but "suggest" is already there and implies the results are not a certainty. I would rather not overly qualify the statement. I have no objections to change if the present statement is not considered NPOV. Ward20 (talk) 19:37, 29 December 2007 (UTC)

After futher thinking about this, I want to read the citation that is used being to support the wording before I jump to any other conclusions. Ward20 (talk) 21:08, 29 December 2007 (UTC)

Maybe it would be better if we just stated that "Cognitive behavioral therapy (CBT) is an evidence based therapy for chronic fatigue syndrome" (I assume no one refutes this) and state nothing about its efficacy in that sentnce. Then we can present the results of all the relevant meta-analyses after (as is currently the case) and let the results speak for themselves. JamesStewart7 (talk) 23:33, 29 December 2007 (UTC)

Or maybe "Cognitive behavioral therapy (CBT) is an evidence based therapy for chronic fatigue syndrome which involves..." JamesStewart7 (talk) 23:35, 29 December 2007 (UTC)

Better not. That would suggest that it actually works. :-) Evidence on CBT is, however, generally poor. Guido den Broeder (talk) 00:14, 30 December 2007 (UTC)

If what is there is not acceptable, the words, "Clinical studies have reported that Cognitive behavioral therapy (CBT) is an effective" might be used. It is not correct to use terms that give a bias either way to the reader. The correct way to address the problem is to have a statement from a WP:RS that describes the evidence base may be flawed. Ref ^[187] is not a full reference if no one knows what that ref is it will be tagged unreferenced. Ward20 (talk) 03:36, 30 December 2007 (UTC)

I would support that Guido how can it claim to be evidence based, when the evidence is flawed? Surely that should be made clear from the outset.

James, this subject is fraught with contradictions as indeed reflected in the science, and sometimes inclusion of a less than ideal study is necessary to show that. The editors here struggle to deal with these issues and less obvious POV than yours. The issues wax and wane between denialism, conservative / establishment dogma, political spin, academic rivalries, commercial opportunism, premature announcements, and frequent easy accusations of fringe science and pseudoscience (rarely validated) are made to support some POV or other. You will need to show more integrity and balance in your edits not just technocrat ability. Let’s hope you can also be constructive?

The mention of the very words Alternative Medicine is enough to invoke knee jerk reactions from some in the extremities! I think I have shown recently, that whether a substance is arbitrarily labelled allopathic or alternative (rather than its use ) will so determine attitudes. When the CDC decides to investigate such practices in CFS and adds its imprimatur to studies of Allostatic Load in CFS; that shifts the balance point of the NPOV position and these become worthy of mention in the Article. Whether such practices will be supported by further study, or explanation of current study remains to be seen. It is to be hoped that the CDC may now divert some funding to this area.

When it is made clear your edits were not welcome, you offer an olive branch, I find devoid of leaves. I was prepared to give you the benefit of doubt with some kind of Freudian slip, if you had corrected further, but instead you plead ‘difficulty’ and demonstrate your word is not to be taken literally. AGP does not seem adequate as now I find on preliminary investigation that you have selectively quote mined from a paragraph the first half of which you quote as if to prove that the second half (that supports a sentence that you have deleted) does not exist! So I have saved you the difficulty and re-instated some missing ‘olive leaves’ to the pre 27th Dec. status (before you added those “long tagged citation needed notices” on the 28th. all in accordance with the discussion. Jagra (talk) 00:34, 30 December 2007 (UTC)

I'm looking at the diff page and all of the reinstated edits are from one section so the good faith assumption would be to assume I just accidentally missed one section. You shouldn't be removing fact tags without explaining how the references support the assertion in the article either. Furthermore if this assertion is not self evident in the reference the fact tag should stay. I believe comments like "this can be explained by..." are OR as the referenced study does not directly refer to the results being explained. This is supported by the wikipedia original research page, "However, using information from references out-of-context or to forward claims not directly supported by the sources is original research."WP:OR Wikipedia policy also supports the tagging/removal of unsourced information " Make sure you supply full citations when adding information and consider tagging or removing uncited/unciteable information in highly controversial articles." As stated here it is perfectly acceptable to tag information before discussing it ont he talk "Any edit lacking a reliable source may be removed, but editors may object if you remove material without giving them a chance to provide references. If you want to request a source for an unsourced statement, consider moving it to the talk page. Alternatively, you may tag a sentence by adding the fact template, a section with unreferencedsection, or the article with refimprove or unreferenced. Use the edit summary to give an explanation of your edit. You may also leave a note on the talk page or an invisible HTML comment on the article page.[2]" [my emphasis]. All the fact tags are going back in. The rest of the edits will be left. Also Jagra, WP:V, quite clearly states the burden of evidence is on you, not me, to support the claim that these pieces of information are referenced "The burden of evidence lies with the editor who adds or restores material.".

So let me see, your logic is that you 'accidently' did not restore that item when you said you would restore all but long standing fact tagged material. But because you tagged it at the time of your revert, those are somehow excluded? Now because someone else reverses for you it is their responsibility, and the burden of evidence lies with them not you? It seems to me you are being devious if not disruptive. Jagra (talk) 04:15, 1 January 2008 (UTC)

Re: evidence based cbt. I was previously unaware that people did not consider the therapys evidence based. Consider this definition though "Evidence-based medicine (EBM) is an attempt to more uniformly apply the standards of evidence gained from the scientific method to certain aspects of medical practice. Specifically, EBM seeks to assess the quality of evidence[1] relevant to the risks and benefits of treatments (including lack of treatment)." If the usage of the treatment and determination of efficacy is primarily based on the application of the scientific method then the treatment should be considered evidence based. Evidence-based does not imply efficacy. Also the standard of evidence for these trials would meet this criteria "Level II-1: Evidence obtained from well-designed controlled trials without randomization." This btw is the same standard of evidence that is required for all surgeries. Surgeries do not have placebo control groups either. You cannot ethically cut someone open for the sake of creating a placebo. So exactly what standards does something have to meet to be considered evidence based. It looks like people are suggesting surgeries are not evidence based. JamesStewart7 (talk) 03:18, 30 December 2007 (UTC)

The 2006 review by Chambers e.a. (but not the abstract) clearly points out some of the weaknesses of CBT studies so far, which include selection bias, small samples, and measurement problems. In addition, there is the large Belgian RIZIV evaluation study of 2006 (in Dutch) that shows that the approach of the Belgian CFS reference centers, which heavily relies on CBT, may improve subjective feelings but does not reduce disability. Guido den Broeder (talk) 11:33, 30 December 2007 (UTC)

I still maintain that evidence based does not imply effective (evidence based simply means the determination of effectiveness is based on evidence) but I'm just going to suggest other options. It seems the primary objection to the term evidence based is a perceived implication of effectiveness so we could use "Some researchers have concluded that, on the basis of clinical studies, Cognitive behavioral therapy (CBT) is an effective, evidence-based therapy for CFS.[reference] Other researchers have stated that there is insufficient evidence to support the effectiveness of CBT [reference]." Other options for the second sentence include "Other reserachers state that the quality of evidence in this area is poor." Hopefully that phrases the CBT debate as an issue of differing scientific opinions as opposed to now where this sentence makes all the research sound like trash "however, the report itself has been criticised by several groups[weasel words], for: being poorly conducted, misrepresentations, omissions, lack of references, factual inaccuracies or bias, and even potentially damaging implications" without actually saying what the problems with the report are. Even if the research is trash no one should be accusing "poorly conducted" for example, without saying exactly why the study was poorly conducted. A better (hypothetical - these will be made up criticisms that should be replaced with the actual criticism) option may be "This study has been critized by John Doe [or X organisation] for lacking patient randomization, using what John Doe beleives to be an overexapansive definition of CFS and poor reporting of drop-outs." With such specific criticisms it is actually possible for the supporters of the reserach to reply. Conversely no one can argue with "poorly conducted" if they don't know what that means.

Re: Belgium study. I placed a verify source tag on that study because the title and presumeably the article is not in English which makes it very difficult to check the study design. Currently the article states "objective measurement shows no reduction in their disability" as I am sure the study did not use every objective measurement on the face of the planet, I would like to see "objective measurements" changed to whatever objective measurements the study used. Also what was the study design and who were the controls?

[5] this review from 2007 states that CBT "appears to be ineffective in many patients". Happy new year everyone! JayEffage (talk) 16:17, 31 December 2007 (UTC)

I would like to point out that virtually every treatment is ineffective in some group of patients eg Penicillin may produce an adverse reaction instead of saving someone. A treatment is deemed to be effective if it produces a statistically significant difference in disease scores. It is not necessary to help everyone for a treatment to be deemed effective. What really matters is how many people are not effected and why the treatment helps some people not others. Note that the researchers seem to imply that CBT is effective for some people. I'm also fairly confident they base the statement CBT "appears to be ineffective in many patients" off another study that they reference as they are not conducting a meta-analysis.

They state that CBT is ineffective in many patients becuase they are speculating the existance of CFS subtypes "Identification of subtypes may lead to more effective therapeutic interventions." I use the term "speculate" because they seem to be basing this conclusion largely on the patient response to CBT therapy. There may be many other reasons for this response. They may have other evidence but I cannot find the full text of the article (maybe it hasn't been published yet) in any journal database I have access to. This review may be worth including if anyone can access the full text. However, it may be better to include this text in the diagnosis or proposed causes section as the article is not about CBT per se. JamesStewart7 (talk) 04:34, 1 January 2008 (UTC)

proposal

This is the finished proposal for alternative medicine:

Two studies reported increased utilization of alternative treatments by people with CFS vs. without. In one study designed to assess use of alternative medicines and perceived benefits, of 126 participants (CFS and non-CFS co-twins), 91% of twins with CFS, and 71% without CFS used at last one alternative treatment in their life, and a large proportion of all twins considered them helpful. Future research should ascertain the usefulness of alternative practices in the management of CFS.^[3][4] Ward20 (talk) 03:16, 30 December 2007 (UTC)

verify source and request quote

(request quote) has been placed on this "According to the Countess of Mar (panel member of the Group on Scientific Research into ME), the report was a political inquiry into the science, not a scientific inquiry, of CFS". My reading of the source is that they were referring to their own convention, not the Gibson report.

(verify source) is placed here "In the early 1990’s Martinovic carried out a clinical trial on CFS patients in which such factors were controlled for and many became fit for work, after about 4 months, and were still well 16 months later" because as I quoted above Martinovic is referring to unpublished research in an article about his hypothesis. If the research is unpublished I don't see how we can verify it.

As I alluded to above, these types of tags should not be removed without discussion. Once people are able to demonstrate that the source supports these statements the tags will be removed. JamesStewart7 (talk) 03:41, 30 December 2007 (UTC)

Whilst the hypothesis says (unpublished work) this is qualified in the conclusion which says “Details of the mandatory dynamic functional assessment will be covered in a subsequent paper” This was correct at the time of acceptance of the paper in Aug. 1993, but by the time of the hypothesis publication with trial results in July 1994, it seems the trial with those details was also published (coincidentally about the same time) in The CFIDS Chronicle, Summer 1994, pp 18 – 28 in a paper titled ‘EFAMs in the pathogenesis and Management of CFS’ by the same Authors. In addition this paper mentions a further paper accepted by the Royal Australian College of General Practitioners as a project under vocational registration requirements, detailing the same trial. (They could hardly have successfully submited at that time a paper considered as ‘alternative medicine’ or even CAM to such a mainstream august body controlling their registration!)

These are not the only publications relating to this trial. In 1998 at the Sydney CFS International Conference, the same Authors presented a lead paper to the multiple CFS fatty acid findings of the Australian University of Newcastle. Their paper was titled “Essential Fatty Acid Metabolite (EFAM) Model of Chronic Fatigue Syndrome (CFS) Management Validated by Successful Predictions: Future Directions” Which not only detailed their predictions validated by the Newcastle papers that followed, but discusses both the two previous papers and provides additional details. They also gave an analysis of controlled CBT trials to that date, differentiate between CBT and CT plus BT and explained their use and assessment with the latter. Explained their interventions and assessment relating to, exercise and their lipid EFAM treatments. They explained why the static dose regime of other EFA trials had reproducability problems, and explained further the factors they controlled for in their trial. Presented again the results from the CFIDS paper and detailed the controlled trials (by others) for separate aspects of their adjuvant interventions in their trial. Showed why in submissions the NHMRC quality of evidence rating for their trial was catergorised as Dramatic results in an uncontrolled experiment. Went on to detail their call for basic science research, and clinical trials and to show how their predicitions had been validated to date.

So it seems the trial details were published on several ocassions. Parameters for assessment of patient outcomes are given in the papers and so the redundent statement can now be removed. The statement including the words “and many became fit for work, after about 4 months, and were still well 16 months later.” Is supported by the words “20 of the 26 compliant patients who had been unfit for full time duties prior to treatment, became fit for full time duties in an average of 111 days (about 4 months) after commencing treatment. Those 20 patients had been on average unable to perform full time duties for a period greater than 3 years prior to treatment At follow up average 16 months later 27/28 were still improved compared to pre-treatment and 20 had added improvements cf 3 months” This appears in the Hypothesis, the Conference paper and among more detailed figures in the CFIDS paper.

The rest of the sentence “In the early 1990’s Martinovic carried out a clinical trial on CFS patients in which such factors were controlled for” is supported by statements in the CFIDS and the Conference papers. In the CFIDS paper “Controlling for these variables not only resolves reproducibility issues, but allows more potent immuno-modulation by titrating a more complete range of FA’s and EFAM modulators.” In the Conference paper; “adjuvant interventions titrated according to changes in symptom subsets and objective monitoring.” And “We predicted a far greater and more consistent incidence of benefit when these (factors described above) were not just controlled for, but dynamically adjusted - as part of our treatment approach.”

The sentence “Cognitive behaviour (CB) was used to show patients (make cognitive) how symptoms varied with activity and dietary changes” is contrasted to statements in the CFIDS paper: ‘In our protocol we routinely applied behavioural modification, low level cognitive therapy, and insight techniques.’ In the Conference paper it distinguishes CT from CBT defining “Cognition - - in this context is the ability to know or perceive”. And “CT - - targetting those perceptions that may aggravate or produce illness.” And also says “we outline that we use in addition to CT other psychotherapeutic processess such as insight techniques” In view of this it is proposed that the sentence be revised to; Cognitive therapy (CT) and insight techniques were used to help patients understand both the nature of their illness and how modulating aspects of treatment might improve or aggravate their condition.

The findings of essential fatty acid deficiencies and explanation of the consequences to CFS symptoms and signs is not included elsewhere in the Article. By default this is included as introductory material in the treatment section on EFAs This needs to be improved and I think I can do it and reduce the words. Consideration can then be given as to where the material belongs. In the meantime I will add further citations, when I can access the Article. Jagra (talk) 10:33, 3 January 2008 (UTC)

Well I've removed the details. If you want to revert pull up the actual publication of the study in a peer-reviewed journal. The researchers claiming the results in multiple papers is not the same as publishing the study. If they didn't publish their study in full in a peer reviewed journal there is no way anyone can verify it and we treat it as Dramatic results in an uncontrolled experiment as the NHMRC classified it. Of course an uncontrolled experiment is a pretty damn low standard of evidence and should just not be included when there are controlled studies available. You shouldn't need to mention 5 different publications that mention the trail or make reference to it or that are not scientific sources. You should just be able to point to one paper that is a published lab report of the original study. In short, don't revert unless you find a publication of the original study that includes methods, results etc reported to an appropriate scientific standard. JamesStewart7 (talk) 14:01, 3 January 2008 (UTC)

I am retracting my request quote over the Gibson report. I initially thought the speaker (Countess of Mar) was unaffiliated with the Gibson Report (I didn't read the intro very well). Turns out I was wrong. I'm actually beginning to question why the Gibson report is included at all. Is the UK parliament really an authority to speak on scientific issues such as the treatment of CFS. Isn't this kind of like quoting George Bush when discussing research on evolution? Why are we allocating so much of the discussion (especially in the GET section) to the Gibson report when there are systematic reviews in this area? The Gibson report does not seem like a reliable source for scientific topics as the members of the discussion are not scientists and it is not a scientific publication. So how would people feel about removing the Gibson report from the article or at least moving it to a "politics" section or similar? I also retract any statments suggesting that it is unwarranted to offer criticism of the Gibson report from the ME association. It is completely appropriate as they are both largely political organisations. The extensive criticism, however, leads me to believe that many editors feel the Gibson report is a rather unreliable source. Given that teh article already has length issues, wouldn't it make more sense just to remove both the report and the crticisms of it? JamesStewart7 (talk) 08:14, 5 January 2008 (UTC)

All references to the Matrinovic trial are now gone. Reinclusion requires a RS for this original study (the actual study itself, not just a reference to the results). Note that a WP:RS for science is one that is peer reviewed Wikipedia:Reliable_source_examples#Cite_peer-reviewed_scientific_publications_and_check_community_consensus so the CFID paper doesn't count. The hypothesis papers doesn't count either as the full methods and results are not published in that paper. JamesStewart7 (talk) 04:38, 6 January 2008 (UTC)

alternative medicine 2

"The mention of the very words Alternative Medicine is enough to invoke knee jerk reactions from some in the extremities" Maybe so but the fact that these pages Orthomolecular medicine (refers to fatty acid supplementation), [acupuncture] include the NCCAM alternative medicince classification template and state things like this "Orthomolecular medicine is practiced by few conventional medical practitioners.[2][3] Orthomolecular treatments are instead more common in complementary and alternative medicine fields, increasingly being integrated into over the counter retail products, naturopathic medical textbooks and mainstream pharmaceuticals" suggests such classification is accurate. The NCCAM even defines fatty acid supplementation as alternative medicine: http://nccam.nih.gov/health/backgrounds/biobasedprac.htm. Seeing as the institute of medicine states "NCCAM has developed what the IOM calls "[o]ne of the most widely used classification structures"" [complementary and alternative medicine], I think NCCAM would qualify as a reliable source. In light of this evidence, I think "Essential Fatty Acid Treatments" and "Magnesium" should be placed under the heading alternative medicine. The current heading structure suggests that these forms of treatments are not alternative medicine (alternative medicine is a heading but these treatments are placed elsewhere). Currently this page is out of harmony with several pages that cover these forms of treatments (listed above). Wikipedia should strive to maintain some degree of consistency beteween articles and the classification structures used in the other articles is very well referenced. JamesStewart7 (talk) 04:54, 30 December 2007 (UTC)

Would calling such treatments experimental take some heat off the discussion? They don't seem to be fundamentally alternative. Guido den Broeder (talk) 11:39, 30 December 2007 (UTC)

Not really. Calling them experimental would still conflict with the other wikipedia pages. Experimental suggests an adherence to the scientific methodology. The concept of Qi , found in acpuncture (just as one example) is not testable so cannot be scientific. More problematically when a physician offers a conventional treatment classified as experimental they are required to call it "experimental". I highly doubt that these treatments (I am referring primarily to acupuncture and orthomolecular medicine btw) are presented to all patients as "experimental treatments". The presence of some scientific studies does not make a treatment experimental. There is the odd study in most areas of alternative medicine. If someone disagrees with my suggestion I really think they need to explain why the NCCAM guildlines are not sufficient to support this categorisation. Also if anyone believes this, they should really also state their complaints on the acpuncture article, the Orthomolecular medicine the complementary and alternative medicine article and the alternative medicine article as these articles make extensive reference to this source. This source is even featured in the alternative medicine template. This doesn't seem like a decision the wikipedia community would make without prior discussion. In fact I can see a reference to that source here [[Talk:Acupuncture#NCCAM_classifies_acupuncture_as_an_energy_therapy]. I can only conclude that some consensus about the reliability of this source has already been reached.

Here is a relevant quotes for the classification by NCCAM "The CAM domain of biologically based practices includes, but is not limited to, botanicals, animal-derived extracts, vitamins, minerals, fatty acids, amino acids, proteins, prebiotics and probioticsLive bacteria (and sometimes yeasts) found in foods such as yogurt or in dietary supplements., whole diets, and functional foods." Also there is a page devoted to just acupuncture "http://nccam.nih.gov/health/acupuncture/" I do not know why this page would be there if they did not consider acupuncture CAM. JamesStewart7 (talk) 13:25, 30 December 2007 (UTC)

I am not referring at acupuncture. Acupuncture can hardly be called experimental after thousands of years. Guido den Broeder (talk) 13:42, 30 December 2007 (UTC)

So are you referring to just orthomolecular medicine (fatty acids, mineral supplementation) or something else completely? I still think the NCCAM page is pretty clear on the classification there. JamesStewart7 (talk) 22:57, 30 December 2007 (UTC)

A few points: I think it is clear from this discussion the Alternative medicine section should be a least be renamed Complementary and alternative medicine. The magnesium section needs a rewrite, the study that showed the most efficacy is not even mentioned. The only sources (as has been pointed out) are some very old studies predating CFIDS. However, another point about magnesium is that it probably should be included under medical treatments. The study that showed efficacy was intramuscular magnesium. Treatment of mineral and vitamin deficiencies have been conventional medicine for a long time. Is it now considered alternative treatment when a person is treated with vitamins and minerals for simple anemia, or does it have to be treated with erythropoietin to qualify as conventional medicine? If I get IV fluids for dehydration is that alternative treatment? Are parts of my chem panels alternative? This all seems rather silly but I am being serious as it seems to me that rigidly subordinating various treatments because of changing and somewhat arbitrary categorizations is somewhat POV and something to be avoided. Similar, but not so clear cut arguements can be made for essential fatty acid supplementation. Cognitive-behavioral therapy is Mind-body medicine and was considered CAM not long ago "Mind-body medicine uses a variety of techniques designed to enhance the mind’s capacity to affect bodily function and symptoms. Some techniques that were considered CAM in the past have become mainstream (for example, patient support groups and cognitive-behavioral therapy)." Ward20 (talk) 01:05, 31 December 2007 (UTC)

Alternative medicine includes the use of treatments that are scientifically proven in specific contexts, in totally separate contexts and for totally separate conditions. Niacin is an excellent treatment for niacin deficiency, but it's an alternative treatment for schizophrenia. It's the use, not the substance, that makes it alternative. Is the use of essential fatty acids in parental nutririon, for neonates,in infant formulas and for conditions VLCAD, LCHAD and TFP deficiencies among others considered alternative, I think not? In this instance Magnesium is advocated because of established medical reason and Essential fatty acids because of identified deficiencies in CFS, neither is an alternative use. The same would apply to other substances found deficient in CFSJagra (talk) 05:10, 1 January 2008 (UTC)

CAM is an acceptable title, however, the fatty acid section should be placed under this title.Both having a CAM title and not placing the supplmentation under it suggests that it is not CAM. If you are not happy with the CAM classification and think that it is POV, I suggest you direct your attention to the orthomolecular medicine page. There has already been discussion of this Talk:Orthomolecular_medicine#Complementary_and_alternative_medicine Furthemore you should look at this page Template:Biologically_based_therapy. Also the complementary medicine page uses the quote you used above and the complementary and alternative medicine page makes extensive definition of CAM. What I am trying to say is that this is not really the place to have this discussion and this is not really the first such discussion to occur on wikipedia. I suggest we use precedent and harmonise this page with the other wikipedia articles and you discuss your proposals on other pages. By discussing such proposals here instead of those pages we would be exlcuding many wikipedians that may be interested in such discussion.

Also re: magnesium efficacy.The results are less than compleeling. Later reviews suggest that the evidence is insufficient to support its effectivness "http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1117488" Actually this review should be added to the article. It states "Subsequent studies have failed to find a deficiency of magnesium in people with chronic fatigue syndrome.27–29". Similar results are reported here http://occmed.oxfordjournals.org/cgi/reprint/55/1/32. Also look at the magnesium RCT. They state that patients were recruited from the centre for the study of complementary medcicine and from GP's. Kind of strange for something that is apparently not part of CAM. In any case, we need to report the whole story for magneisum. The other side of the story being that magnesium levels may not actually be lower in people with CFS. Magnesium may be considered not complementary when we have more than one small RCT on the topic or when there is some medical rationale for giving magnesium. Currently there is still a strong possibility that the patient selection criteria biased the results (it was reported in the same paper that CFS patients have a magnesium deficiency when other studies found otherwise). With essential fatty acids the evidence is even less convincing. All we have is one study that could not be replicated and one case where a researcher claimed that essential fatty acid treatment worked but never published his results. This complete lack of replication is exactly why these treatments are still CAM. JamesStewart7 (talk) 05:34, 1 January 2008 (UTC)

Would you give a link to the section where there has already been discussion of this on Talk:Orthomolecular_medicine#Complementary_and_alternative_medicine, so I know we are looking at the same discussion? Thanks. Ward20 (talk) 20:45, 1 January 2008 (UTC)

Here is one such discussion http://en.wikipedia.org/wiki/Talk:Orthomolecular_medicine#.22Complementary_and_alternative_medicine.22. Here are two other relevant discussions http://en.wikipedia.org/wiki/Talk:Orthomolecular_medicine/Archive_3#.22Evidence-based_CAM.22.3F, http://en.wikipedia.org/wiki/Talk:Orthomolecular_medicine/Archive_3#Woah.21_Orthomolecular_medicine_is_not_the_same_as_alternative_medicine. Here is yet another, http://en.wikipedia.org/wiki/Talk:Orthomolecular_medicine/Archive_2#.22alternative_medicine.22. Here is another where someone complains about CAM being moved from the intro http://en.wikipedia.org/wiki/Talk:Orthomolecular_medicine/Archive_2#added_factual_accuracy_and_pov_tag. There may be other discussion on other pages (perhaps the alternative medicine or the CAM template) page. JamesStewart7 (talk) 05:45, 3 January 2008 (UTC)

So would people agree that we should reviews for the essential fatty acid and magnesium sections instead of original studies as they are secondary sources (see Wikipedia:Primary_sources#Primary.2C_secondary.2C_and_tertiary_sources) JamesStewart7 (talk) 07:09, 3 January 2008 (UTC)

No, because: 1.) The reviews that have been put forth were written by well known CBT advocates. 2.) The studies used different inclusion criteria, different testing methods and came to different conclusions. The reviews do not give this any weight and are written like opinion pieces, not reviews.

This is an example of the bias: Summary; "One small RCT found limited evidence of benefit from magnesium injections." Detailed description of study; "This trial found important benefits with magnesium"...

Reviews should be found by neutral sources, or the studies should be talked about in the article based on the inclusion definition, the methods used, and the conclusions. Ward20 (talk) 08:02, 3 January 2008 (UTC)

No, Reveiws can be useful to clarify say a foreign paper with a limited abstract, but can also introduce their own bias when they generalise, and can skate over what may seem trivial points today, but be key to tomorrows perspective, better for editors to determine on a case by case basis. Jagra (talk) 10:43, 3 January 2008 (UTC)

So suddenly journal articles from the British Medical Journal are not a reliable or neutral source? Reviews are meant to synthesize data and draw conclusions. I do not agree that this is bias "Summary; "One small RCT found limited evidence of benefit from magnesium injections." Detailed description of study; "This trial found important benefits with magnesium"..." One small RCT is limited evidence. It doesn't matter what the area is. You cannot make solid conclusions with just one study, especially when other papers refute aspects of it. Also they are well known CBT advocates? Do you have a RS for this? Do you have a RS that demonstrates that their opinon is not based on the scientific data? Lets not pretend anyone can objectively evalute what is and is not a POV source which is the rationale between wikipedia NPOV policy (representing all viewpoints in proportion to the reliable sources) so I don't think merely calling a source biased is an accpetable reason for excluding it.

Correct me if I'm wrong but doesn't virtually every other wikipedia article use reviews and meta-analysis if they are available. Using published reviews and meta-analyses prevents reviewers from doing this Wikipedia:Primary_sources#Synthesis_of_published_material_serving_to_advance_a_position. With original studies you can pick those that support your position, ignore the negative results found in other studies and then state that your position is supported. WIth a review/meta-analysis you cannot do this as reviews and meta-analyses must cover a broad range of studies. If youd don't like the conclusion stated in the review, find another review. I am quite happy to include every review on this topic that is published in a medical journal. We are not here to decide what is a good scientific conclusion and what is a bad (and consequently POV) scientific decision, that is OR. We're supposed to refer to the reliable sources (the decisions the scientists made and let the facts stand. Why should we be synthesizing scientific material and drawing our conclusions from that when scientists have already done it? Since when is OR preferrable to a secondary source?

The reviews should be included. "The threshold for inclusion in Wikipedia is verifiability, not truth". The reviews are clearly a reliable source. Not liking their conclusion is not a good enough reason to exclude them. They are a reliable source published in a prominent journal. Please find some problem with the reliability of the source if you do not think it should be included. This "and can skate over what may seem trivial points today, but be key to tomorrows perspective," violates Wikipedia:Crystal#Wikipedia_is_not_a_crystal_ball. Also note WP:NPOV policy "NPOV says that the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each". This means that if there are more reliable sources (which I guess would be journal articles that review the effectiveness of these treatments as a whole ie a review or meta-analysis) that state that magnesium/essential fatty acid treatment is effective/ineffective then this viewpoint should receive the most consideration. The other viewpoint should also receive some consideration if there are reliable sources stating that viewpoint (of course only scientific sources can be considered reliable sources of scientific opinions). JamesStewart7 (talk) 14:21, 3 January 2008 (UTC)

So suddenly journal articles from the British Medical Journal are not a reliable or neutral source? James Stewart

As far as ME/CFS studies are concerned, the answer is a firm yes. With I think just one exception, this otherwise reputable journal has published only behavioural based studies for about 15 years. An outstanding example of publication bias if ever I have seen one.

See Goudsmit E and Stouten B. Editorial bias in the British Medical Journal for a critique of the period covering 1995-2000. Bricker (talk) 16:52, 3 January 2008 (UTC)

It doesn't seem like this is a standard stuy of publication bias "Analysis of the findings revealed a bias towards the views of one school of thought and a lack of papers on the immunological or virological aspects of CFS". This is not pulbication bias. "Publication bias arises from the tendency for researchers and editors to handle experimental results that are positive (they found something) differently from results that are negative (found that something did not happen) or inconclusive." It looks like what they are discussing is something else entirely. I'm sorry, we can't just throw out all the studies published in one journal because they report one school of thought more than another one. If a biology journal reports on evolution more than intelligent design does that make it biased? Even if we are to believe that the BMJ is biassed it is still a reliable source, being a high impact scientific journal. Also if you don't like the statement giving in the BMJ article review go ahead and find another one. I'm happy to report the conclusions made in all reviews, and I think high impact journals should reasonably be given more consideration as impact factor is a measure of importance and therfore reliability. JamesStewart7 (talk) 23:18, 3 January 2008 (UTC)

Also the Journal of Chronic Fatigue does not seem to follow a standard peer review process "While it is important to publish in many different peer-reviewed journals, expert reviews take up lengthy periods of time and the lag time between submission and publication can often be up to two years. For that reason, a well-respected expert panel has been drawn together to serve as the Editorial Board, which allows for critical, rapid reviews of basic, clinical, and multidisciplinary work." They never actually say it but the implication here seems to be they abandon the standard, long peer review process favouring a faster process where only a select few can review the material. Doesn't really lend to the credibility of this journal does it? JamesStewart7 (talk) 23:36, 3 January 2008 (UTC)

James, a review of a few of your arguments: "Lets not pretend anyone can objectively evalute what is and is not a POV source which is the rationale between wikipedia NPOV policy (representing all viewpoints in proportion to the reliable sources) so I don't think merely calling a source biased is an accpetable reason for excluding it." And, "We are not here to decide what is a good scientific conclusion and what is a bad (and consequently POV) scientific decision, that is OR."

Then, "the implication here seems to be they abandon the standard, long peer review process favouring a faster process where only a select few can review the material. Doesn't really lend to the credibility of this journal does it? " is WP:OR.

And the argument, "However, I wouldn't call this the most reliable source. THe source was published in integrative medicine, a journal which survived only three years and published articles such as" The Plausibility of Homeopathy The Systemic Memory Mechanism" in which it was stated that homeopathy is highly plausible homeopathy however, provides the complete oppostive view and is well referenced). Due to the presence of articles such as this in the journal, I consider the source itself, POV." is WP:OR and counter to your statement above.

Then, "It doesn't seem like this is a standard stuy of publication bias" is WP:OR.

I agree, we should try to stick to guidelines and policies of WP:RS, WP:V, and WP:OR, and I think there should caution and fairness about it in this WP:Controversial article whatever POV the editor has. Ward20 (talk) 02:37, 4 January 2008 (UTC)

Yeah my OR is on the discussion page. Not every argument I state here is something I want to be directly included into the article. For example, this was obviously not intended to be included in the article "Lets not pretend anyone can objectively evalute what is and is not a POV source which is the rationale between wikipedia NPOV policy (representing all viewpoints in proportion to the reliable sources) so your argument is somewhat of a straw man. I am objecting to OR which is in the actual article. The synthesis of scientific studies and drawing conclusions from multiple studies (which are often contradictory) is OR. Using the reviews is not OR. It is using a reliable secondary source. There is no way anyone can tell me that the British medical journal is not a reliable source given its prominence so I think you would be hard pressed to exclude it on reliability issues. Seeing as relialability and verifiability is the only criteria for inclusion in wikipedia "the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each", "The threshold for inclusion in Wikipedia is verifiability, not truth". Refusing to include a reliable secondary source because you do not like the conclusions they make is a clear violation of wikipedia policy. If anyone really thinks the British medical journal is not a reliable source I suggest we open a RfC and see what the wider wikipedia community has to say. JamesStewart7 (talk) 04:46, 4 January 2008 (UTC)

It seemed to me your point was that myself and others arguing about review articles in the British Medical Journal being biased and a reason to use the primary sources rather than secondary sources was WP:OR (and it may be), and should not be done. Then it appeared you were making similar arguments about the reliability of review articles that others proposed as sources. If I mis-characterized your position I apologize. I did not note anyone saying the BMJ is not WS:RS. Only that some studies advocating CBT and GET as superior to other therapies in the BMJ are flawed, and are stated as such in other WS:RS.

Just a few other examples:[6]"Criticisms of the Royal Colleges Report: The report drew a fiercely critical response. Commentators50 have suggested that the working group was heavily weighted with psychiatrists, giving it a natural bias towards psychiatric rather than physiological explanations for the condition.51 An editorial in the Lancet endorsed the view of patient groups that the report was too ready to dismiss for lack of evidence a major role for viral cause of CFS or for structural or functional abnormalities in muscle or brain. It states the report was 'biased and inconclusive'.52 These criticisms were rejected by the Royal Colleges. The Lancet's article was condemned, in a letter to the Lancet, as 'misleading and tendentious', and the report of the working party strongly defended:53"

Dr. Philip Lee Assistant Secretary of Health, US DHHS 1993-1997 Acceptance Speech[7] for the Rudy Perpich Award presented at the Bi-Annual Research Conference of the American Association for Chronic Fatigue Syndrome (AACFS) October 10-11, 1998 Cambridge, Massachusetts: "Second, the approach to CFS is now dominated by the biopsychosocial approach that gives excessive emphasis to the social, behavioral, and emotional factors in the presentation and perpetuation of symptoms. The "bio" seems to be missing. While I believe in the psychosocial determinants of health paradigm, this approach to CFS has gone too far.

The problem is evidence in the proposed ICD-9 codes for CFS, and the 1996 report of the Joint Working Group of the Royal Colleges of Physicians, Psychiatrists and General Practitioners on Chronic Fatigue Syndrome in the United Kingdom. The Royal Colleges convened a working group after a request from the UK's Chief Medical Officer. The group recommended that the term encephalomyelitis be dropped in the UK and that it be replaced by CFS."

That is just the tip of the iceberg. I was going to suggest something similar to yours. We post the question of how to best incorporate WP:RS criticism of reviews or studies from a top quality WP:RS journal without ending up with confusing arguments. I would have recommended the Wikipedia:Reliable sources/Noticeboard as they seem to be focused. The arguments might be roughed out on the talk page first and then posted there or at Wikipedia:Requests for comment.

I intend to finish editing the CFS paragraph I have been working on tomorrow, describe some potential review sources on the talk page, and take a long break. If that discussion occurs, I probably will not participate. Ward20 (talk) 09:52, 4 January 2008 (UTC)

Note that if a journal is considered a reliable source, this does not automatically imply that every article published in it is also reliable. Authors, as well as peers, make errors. It happens often enough that the reliability of an article is contested in a new article or a letter in the same journal, or in another journal that is also considered a reliable source. We therefore need not discuss the reliability of the BMJ in general. Guido den Broeder (talk) 12:12, 4 January 2008 (UTC)

Ok well what objective standard of reliability would we use then? WP:RS infact suggests some objective criterion that may be used:

Scholarship

Wikipedia relies heavily upon the established literature created by scientists, scholars and researchers around the world. Items that fit this criterion are usually considered reliable. However, they may be outdated by more recent research, or controversial in the sense that there are alternative scholarly explanations. Wikipedia articles should point to all major scholarly interpretations of a topic.

The material has been thoroughly vetted by the scholarly community. This means published in peer-reviewed sources, and reviewed and judged acceptable scholarship by the academic journals.

Items that are recommended in scholarly bibliographies are preferred.

Items that are signed are more reliable than unsigned articles because it tells whether an expert wrote it and took responsibility for it.

So apparently peer reviewed sources are considered reliable. Now all articles published in the BMJ are peer reviewed. The journal of chronic fatigue syndrome is not peer reviewed therefore. Therefore BMJ articles are more reliable than Journal of CFS articles. As wikipedia policy states "Wikipedia articles should point to all major scholarly interpretations of a topic", it seems logical that we include all reviews in this area. I actually suggested two reviews http://occmed.oxfordjournals.org/cgi/reprint/55/1/32 and http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1117488. The problem with all these alternative treatments as the Occupational Medicine article suggests is "Most of the

other interventions were evaluated in just one or two studies and therefore evidence is insufficient to draw firm conclusions". The BMJ article assumes a similar posiition. So we have two reviews, conducted by two groups of people (although Chadler participated in both reviews) stating that there is not enough evidence to support the effectiveness of magnesium or essential fatty acid treatment. As far as I am aware, no retraction has been placed on either of these articles. As wikipedia policy states "Wikipedia articles should point to all major scholarly interpretations of a topic" we should refer to all scholarly interpretations in this area. A wikipedia editor concluding that one study is enough evidence to support the effectiveness of a treatment is not a scholarly interpretation. This is also a scholarly interpretation (although it doesn't actually deal with supplement treatment which is the content of this discussion) but being a consensus statement for a group of doctors, should be attributed to the source (Royal College of Physcians). An editorial in the Lancet is also a scholarly opinion but being an editorial and not a research paper, the opinions expressed should be attributed to the author. This also warrants inclusion http://www.cfids-me.org/mpwc/lee.html needs to be attributed to the author. All of these criticisim papers, however, have absolutely nothing to do with essential fatty acid/magnesium treatment. Even if we accept that CFS is a biological condition that doesn't mean that essential fatty acid or magnesium treatment is effective. I have no problem with including these reliable sources suggested by Ward20 as long as the opinions expressed are attributed to the author but I see no reason to include them in the Essential fatty acid/magnesium section as they are completely off topic.

"If I mis-characterized your position I apologize." No, you reasonably characterized my position. I had (and still have two main points). The BMJ article and Occupational medicine reviews are reliable sources as they are peer reviewed and attributed to their author and secondly, review articles are preferrable to to original studies as wikipedians often have to make their own interpretations when synthesizing a group of scientific studies (which is OR). "Only that some studies advocating CBT and GET as superior to other therapies in the BMJ are flawed, and are stated as such in other WS:RS" This is not really relevant to what I was saying (I was not doing a comparison of CBT/GET therapy to other treatments) but I will address this point anyway. GET/CBT are the only two therapies on which an extensive amount of reserach has been done (appart from maybe anti-depressants but I haven't really reviewed that area). The review articles in question did not simply state that CBT/GET are effective for everyone. They instead used words such as promising/effective for some. The meta-analyses of CBT/GET are also more reliable than these review articles. For this reason, I wouldn't even include their opinions on CBT/GET. Meta-analyses are more reliable when there are large number of studies as you have to clearly state your inclusion/exclusion criteria and weight all the studies accordingly. With reviews you can pick and choose. Reviews, however, work well when there are only about 3 studies in the area. The systematic review referenced in the article, however, didn't exist when these reviews were being conducted so their approach was reasonable. So if anyone thinks I am suggesting that we include these reviews as support for CBT/GET treatment they have misinterpreted me. I am suggesting that we use these reviews for magnesium/essential fatty acid treatment only as no systematic-reviews (meta-analysis) exists in this area as there are not enough studies to make it worthwhile. JamesStewart7 (talk) 04:43, 5 January 2008 (UTC)

Also thanks for providing the link to the reliable sources noticeboard. I wasn't previously aware of this board. I will post a comment there if people think it is worthwhile (ie enough people think the BMJ review is not a RS). JamesStewart7 (talk) 04:46, 5 January 2008 (UTC)

I was also looking at the wikipedia source list and found this Interventions for the treatment, management and rehabilitation of patients with chronic fatigue syndrome/myalgic encephalomyelitis: an updated systematic review. I am having trouble getting the full text of this article but they state "Studies on behavioural, immunological, pharmacological and complementary therapies, nutritional supplements and miscellaneous other interventions were identified. Graded exercise therapy and cognitive behaviour therapy appeared to reduce symptoms and improve function based on evidence from RCTs. For most other interventions, evidence of effectiveness was inconclusive and some interventions were associated with significant adverse effects." As they only concluded that behaviour therapies were effective I can only assume that they also concluded that there was sufficient evidence to support all the complementary treatments. That makes 3 review articles (http://occmed.oxfordjournals.org/cgi/reprint/55/1/32, http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1117488, http://www.jrsm.org/cgi/content/abstract/99/10/506) that have all reached the same conclusions as opposed to zero that anyone has referenced that have found an opposing conclusion. According to WP:NPOV "NPOV says that the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each" so the article should reflect the fact that three reliable sources have found concluded that there is insufficient evidence for the efficacy of nutritional supplements in the treatment of CFS. So this means 100% of the discussion should cover these sources until someoen can find a RS for the opposing opinion. When a reliable source publishes a different conclusion about the efficacy of nutritional supplements for CFS that source may be included. As far as I am aware no such source exists. It is reasonable as Ward20 suggests, "to best incorporate WP:RS criticism of reviews or studies from a top quality WP:RS journal" as soon as anyone can point to this source. This source, https://www.haworthpress.com/store/ArticleAbstract.asp?sid=4BSB0H1E6U219KEMSDN8UAW28XGW2577&ID=62768, however, would not be appropriate as it a)states absolutely nothing about the effectiveness of nutritional supplement treatment or the methods used in the reviews and b) it is not a reliable source of scientific opinion as it is not peer reviewed. JamesStewart7 (talk) 12:09, 5 January 2008 (UTC)

I believe you are wrong about the the Journal of Chronic Fatigue Syndrome not being is peer reviewed. It may be an expedited review, but they state it is a peer review process.

• [8] "While it is important to publish in many different peer-reviewed journals, expert reviews take up lengthy periods of time and the lag time between submission and publication can often be up to two years. For that reason, a well-respected expert panel has been drawn together to serve as the Editorial Board, which allows for critical, rapid reviews of basic, clinical, and multidisciplinary work."
• And here "JOURNAL OF CHRONIC FATIGUE SYNDROME™—Instructions for Authors 3. COPYRIGHT. Copyright ownership of your manuscript must be transferred officially to The Haworth Press, Inc. before we can begin the peer-review process. The editor’s letter acknowledging receipt of the manuscript will be accompanied by a form fully explaining this. All authors must sign the form and return the original to the editor as soon as possible. Failure to return the copyright form in a timely fashion will result in a delay in review and subsequent publication. [See MANUSCRIPT SUBMISSION FORM]"

Ward20 (talk) 08:26, 6 January 2008 (UTC)

Ok I'll accept it has a limited peer review process. That still places it below journals with a more extensive peer review process and there is still the problem that the cited article is irrelevant to the effectiveness of any alternative treatments so there is no reason to include it. Besides, there are three papers in three journals that reach the same conclusion and the CFS journal article only refers to the BMJ. JamesStewart7 (talk) 10:03, 6 January 2008 (UTC)

Here is a review on magnesium usage in CFS that goes into great detail. The review also mentions a review from Japan that states, "Trials with magnesium have improved the well-being of patients." Although it is not clear which trials the Japan review is describing. The author discusses possible reasons why one study may have shown positive results and why another did not (the second study only gave a single injection which the Occupational Medicine 2005 review failed to mention). The Occupational Medicine 2005 review also failed to mention only 20% of the participants of the other cited study, that failed to find magnesium deficiency in people with CFS, met the CDC criteria for CFS. Ward20 (talk) 10:38, 6 January 2008 (UTC)

Am I to take it that we are in agreement that reviews should be used to cover magnesium and essential fatty acid treatment and we are just sorting out WP:WEIGHT issues? From what I can see this review never actually says Mg treatment is effective "[9]. Also the paper is speculative eg "POSSIBLE MAGNESIUM-DEFICIENCY MEDIATED MECHANISMS IN CFS AND FM". It also states this "Since there is no satisfactory therapy for CFS, and there is a Mg-responsive condition: LTS-which resembles CFS and FM, the Mg status of CFS patients should be evaluated". The three reviews I gave referred to papers that did evalute this status. If I recall correctly there was one study that found Mg to be lower in CFS patients and two studies that failed to replicate this result. Anyway the point is I don't see what this paper would add as it doesn't discuss the efficacy of the treatment (which is what the section is about) and they are largely just suggesting hypotheses "...which might contribute to the postulated increased..." This is in Japanese [10] so it should not be included. The rest of your comment is your OR so it cannot be considered for inclusion in the article. JamesStewart7 (talk) 11:19, 6 January 2008 (UTC)

Should is the operative word. As detailed as the Seelig review was there were no definate conclusions such as the other reviews made. There were also statements of problems with studies that the other reviews did not mention. Please point out where I made any OR in my arguement. AFAIK I just stated facts included in the reviews or studies. I stated no opinions, extrapolations, or conclusions. If facts show there are problem with the reviews you want to use so be it. WP:RS also gives an external links to How to Read a Primary SourceHow to Read a Secondary Source, . It states, "What does the author say?, Why does the author say it?, Where is the author's argument weak or vulnerable?" I don't think we need to be stupid or blind to a source's bias. We need to find better sources, or show how weak these sources are with other RS viewpoints. Ward20 (talk) 12:58, 6 January 2008 (UTC)

RS is a guideline "and should be treated with common sense and the occasional exception." If using secondary sources will not give the reader information they should have to make an informed decision, and primary sources written in a NPOV manner will, then they should have the information, or IMO we are not writing an encyclopedic article. Ward20 (talk) 21:32, 6 January 2008 (UTC)

Ward20 I think your recent edits are good and I have no problem with these edits. However, I think there are currently weight issues (due to not including the reviews) which I will try to correct. I'm all for "find(ing) better sources, or show how weak these sources are with other RS viewpoints" but I can't read Japanese. I'm going to go out on a limb though and say when the Japanese review says "Trials with magnesium have improved the well-being of patients" they are referring to the same small RCT that was included in the other three reviews. Also please note the Japenese translation does not say that magnesium is effective. It basically just says a few trials showed positive results. We don't even know the Japenese review does offer a differing view. I believe using secondary source will give a reader all the information they need but if you want to include a note from this study "http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=PubMed&dopt=AbstractPlus&list_uids=7738491" then that is fine, however the other suggestions "reasons why __ may have shown" are speculative and may not line up with the source. I think common sense would suggest that we shouldn't make overreaching interpretations from an abstract of a review we can't read. JamesStewart7 (talk) 23:39, 6 January 2008 (UTC)

James, what you say above is sensible. I hope everyone's POV can blend into NPOV text and we incrementally improve the article without grating on each other. As I said before, I think the Mg section needs a rewrite. It would be nice if we could get a translation of the Japanese Mg review. Anyone know a way to contact the Japanese Wiki? Ward20 (talk) 02:28, 7 January 2008 (UTC)

I've changed the classification of essential fatty acid supplementation and magnesium. There is some form of consensus supporting this as the [Template:Biologically_based_therapy]] has included orthomolecular medicine for some time now. Likewise the page orthomolecular medicine has listed it as CAM for months if not longer and explicitly included essential fatty acid treatment and mineral supplementation as part of orthomolecular medicine. Furthermore the WP:RS, NCCAM trumps anyone's opinion on the matter so if anyone reverts they need to have a source more reliable than NCCAM and they should establish the reliability of this source on the orthomolecular medicine and complementary and alternative medicine pages. JamesStewart7 (talk) 04:38, 6 January 2008 (UTC)

Attribution of source

"The name chronic fatigue syndrome is itself controversial, with patient advocates and medical professionals[4] preferring terms such as myalgic encephalomyelitis ("ME" or "ME/CFS") and post-viral fatigue syndrome ("PVFS"), which imply specific underlying etiologies or pathologic processes" As this is a controversial article, I think it would be best if we can attribute statements like these to whoever said them. Ward20 made a positive move by adding another source for the statement but I think the statement above implies all patient advocates and medical professionals prefer such terms. Looking at the source, this quote seems relevant "She [Jill McLaughlin] noted that myalgic encephalomyelitis (ME) is a more specific and appropriate diagnosis than Chronic Fatigue Syndrome. She asked that the Name Change Workgroup be supported to continue their work and that CFSAC hold a name change session." However, the fact she is suggesting this recommendation to the CFSAC suggests they are yet to accept this change..

So it seems to me that the above quoted sentence may be worded as "The name chronic fatigue syndrome is itself controversial, with the National Chronic Fatigue Immune Dysfunction Foundation [4] preferring terms such as myalgic encephalomyelitis ("ME" or "ME/CFS") and post-viral fatigue syndrome ("PVFS"), which imply specific underlying etiologies or pathologic processes" I am going to implement this change now but if anyone believes that the opinions of Jill McLaughin are not representative of the NCF please revert it. Also if anyone wishes to add other persons/organistaions to the list, I have no objections as long as the opinions are attributed to relevant persons/organisations.

It would also be good to see similar changes made in other secitons tagged weasel words so we can remove these tags. JamesStewart7 (talk) 08:33, 30 December 2007 (UTC)

Note the recent name change of the international association of scientists and clinicians from IACFS to IACFS/ME. Regards, Guido den Broeder (talk) 11:43, 30 December 2007 (UTC)

Added text with cites to show the magnitude of non-support by patients in America for the name Chronic Fatigue Syndrome. Ward20 (talk) 11:53, 30 December 2007 (UTC)

Propose moving, "Eighty-five percent of respondents of a survey conducted by the Chronic Fatigue Immune Dysfunction Syndrome Association of America indicated they wanted the name changed.^[5]The association ^[6] prefers terms such as myalgic encephalomyelitis ("ME" or "ME/CFS") and post-viral fatigue syndrome ("PVFS"), which imply specific underlying etiologies or pathologic processes."^[7] to Nomenclature; * Chronic fatigue syndrome (CFS). This level of detail is not needed in the lead, and the second sentence still needs minor work to present the reader with a complete understanding of the issue. Ward20 (talk) 19:03, 30 December 2007 (UTC)

Where would you move it to though? There is no clear section on this in the article. This is not to say there shouldn't be though. WP:Lead states that the lead should do this, "The lead should be capable of standing alone as a concise overview of the article, establishing context, summarizing the most important points, explaining why the subject is interesting or notable, and briefly describing its notable controversies, if there are any." Right now the lead seems to go into fairly extensive detail about what CFS is and touches on diagnosis but it seems to largely miss out any summary of the proposed causes and treatments sections. There are several sentences that go into too much detail eg this may be better placed in the diagnosis section "CFS patients may report many other symptoms which are not included in all diagnostic criteria, including muscle weakness, cognitive dysfunction, hypersensitivity, orthostatic intolerance, digestive disturbances, depression, poor immune response, and cardiac and respiratory problems." Maybe it would be easiest to alter the "controversy over the term CFS" sentence until everyone is happy with it and then maybe consider moving the sentence when we take a look at the lead as a whole. JamesStewart7 (talk) 06:37, 31 December 2007 (UTC)

I would move the two sentences to here:

Nomenclature

to look like this:

• Chronic fatigue syndrome (CFS); this name was introduced non-unanimously in 1988 by a group of United States researchers based at the Centers for Disease Control and Prevention in response to the 1984 Lake Tahoe ME epidemic, and is used increasingly over other designations, particularly in the United States. Many patients and clinicians perceive the term as trivializing^[5] and as the 1994 Fukuda paper itself cedes, stigmatizing which has led to a campaigning movement to change the name and definition. Eighty-five percent of respondents of a survey conducted by the Chronic Fatigue Immune Dysfunction Syndrome Association of America indicated they wanted the name changed.^[5] The association ^[8] prefers terms such as myalgic encephalomyelitis ("ME" or "ME/CFS") and post-viral fatigue syndrome ("PVFS"), which imply specific underlying etiologies or pathologic processes.^[9]

My view is that only the last sentence needs some work but if you want to to wait until it is reworked where it sits in the lead, it is all right with me.Ward20 (talk) 07:19, 31 December 2007 (UTC)

Yeah it probably would be better in that place. I really don't mind whether people move it now or later (either when they are satisfied with the sentence or when we work on the lead). As you have already picked out a place to move it, I see no real reason to wait, provided of course, everyone else is happy with the move. JamesStewart7 (talk) 10:11, 31 December 2007 (UTC)

Per citation needed for, "in 1978 the Royal Society of Medicine accepted ME as a distinct clinical entity.[citation needed]" possible source Epidemic Neuromyasthenia 1934 1977.. current approaches. Ed: WH Lyle and RN Chamberlain. Postgraduate Medical Journal 1978:54:637:705 774 pub: Blackwell Scientific Publications, Oxford^[10] I am not in England, can someone verify?Ward20 (talk) 02:10, 31 December 2007 (UTC)

It's in Ramsay's book. Guido den Broeder (talk) 11:52, 1 January 2008 (UTC)

Wallman et al

I found Wallman et al 2004, "https://www.mja.com.au/~matty/wal10613_fm.pdf". The article stated "The original concept of pacing was tested in a RCT by Wallman et al (2004) where it was found to be helpful, but not effective enough to be considered a form of therapy." Wallman et al, however, state "Our study showed that a graded exercise intervention that included a form of pacing was associated with specific physiological, psychological and cognitive improvements as compared with a control regimen of relaxation/flexibility. An important aspect of our study was that graded exercise was not associated with a relapse in any participant. While not a cure for CFS, graded exercise improves functional ability, and minimises deconditioning, which can result in more symptoms." This leaves two problems with the previous revision of the article. Wallman et al. stated it was not a complete cure for CFS. That is a far cry from "not effective enough to be considered a treatment". Secondly the article is titled, "RCT of GET in CFS" yet the article presented pacing and GET as seperate things although I will concede they were presented as similar.

Wallman et al. 2004 is a reliable source for the GET section but as it is a primary source, the already included secondary sources may be preferrable. The similar/related treatment section, however, was devoid of references so I am going to call WP:PROVEIT on this information. JamesStewart7 (talk) 11:33, 5 January 2008 (UTC)

GET unreferenced statment removed

Also continuing with WP:PROVEIT I removed this statement because it makes several unreferenced implications "However this level of efficacy was only found in several small trials and was not compared with specialist medical care or pacing." Firstly as stated in the edit note, there were 70 studies going into this article. I am sure they were not all related to GET but as I am seeing at 5 studies on GET in this 2005 review, I really doubt the term several (I don't really consider 5 several. Also if it is 5 ). Btw as the Royal Society of Medicine reference was only used for GET/CBT and not other treatments I initially assumed the 70 studies were for CBT/GET. As I now know this to be incorrect, I think that if there is <10 studies in the Royal society of medicine article, these reviews may be considered for inclusion http://occmed.oxfordjournals.org/cgi/reprint/55/1/32, http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1117488 so ignore my previous comment on this. Furthermore this is misleading "not compared with specialist medical care or pacing". Firstly I see no research on pacing (and that other reference to it I removed earlier was GET). Secondly what is specialist medical care and how is this any different from the control groups that were used. Thirdly I see in this article, http://occmed.oxfordjournals.org/cgi/reprint/55/1/32, that GET was compared to CBT and no other effective treatments have been found so why would they compare it to anything else. JamesStewart7 (talk) 12:50, 5 January 2008 (UTC)

I think I should clarify why exactly this statement needs a reference. It is prefaced with "however" which presents it as a limitation of a study so a reference is needed to say that the study was limited in this way. Any limitations that the authors state may also be used. They do give limitations and areas for future research eg "There is a need for research to define the characteristics of patients who would benefit from specific interventions and to develop clinically relevant objective outcome measure". Right now the problem is that the statement reads like an editors criticism. Any criticisms must be referenced. JamesStewart7 (talk) 02:01, 6 January 2008 (UTC)

If you remove 'however' and 'only', you can keep it in as a neutral observation. Guido den Broeder (talk) 10:17, 6 January 2008 (UTC)

The "Gibson Report" page 23/24: The evidence given to the Group by Dr Peter White found that in four studies 50-70% of patients improved with GET[17]. However, Dr White also states that "GET (and CBT) have been shown to be efficacious only in small trials. They have never been compared to specialist medical care or pacing. We do not know the best treatment; for whom; nor how they work"[18] - Tekaphor (talk) 01:55, 7 January 2008 (UTC)

Reliable sources for scientific statements

The following things are not reliable sources for scientific statements

1. Letters to the editor (Neuro endocrinology letters)
2. Conference proceedings (may be suitable for statements of consensus provided the organisation is reputable and the statement is attributed) - it was not
3. Hypothesis papers - they are fine for a statement of a hypothesis but they are largely speculative. They cannot be used to explain differences in conflicting findings. More studies are needed for that.
4. Political proceedings - The Gibson report. I haven't removed this yet but the Gibson report was not conducted by a scientific organisation therefore they have only a political opinon, not a scientific one. The Gibson report is a reliable source for poltical opinions but it has been used in a sceintific context which is inappropriate.
5. Any journal that is not peer reviewed. (Wikipedia has some more guildlines for what journals are reliable)
6. Anything with a reference that has the word "might" in the title. This typically indicates a speculative statement or a hypothesis. These papers are published to outline areas for future research they should not be used amongst evidence supported statements as it easy to accidentally convey the impression that these are evidence supported statements (science commonly uses words like "suggest", "indicate" even when there is lots of evidence). Also this article should really be about outlining the current state of the research, not speculating future research (see WP:Crystal Ball).
7. Unpublished research.
8. Case studies - they may be an RS in some cases but there are big WP:WEIGHT issues as the results are fairly worthless. see Wikipedia:Reliable_source_examples#Evaluating_experiments_and_studies. For this article, where there are reviews and controlled studies available, case studies do not warrant inclusion at all.

Really, the only thing that is a RS for a scientific statment is a review/meta-analysis/original study published in a peer reviewed journal. That is not to say that other sources cannot be included in the article, but it must be made clear that these are not scientific opinions or statements of fact (eg Gibson report is a RS for a political opinion but not a scientific one). Furthermore anything that is not directly supported by the source is OR eg this unreferenced statement "The mixed results of earlier trials was explained as these likely did not attempt to control for other intake or lifestyle factors that effect essential fatty acid utilization." Regardless of how valid this inference may be (personally it is not how I would explain the difference in results) it is still OR.

Hopefully by now it should be clear why I recently performed a revert (btw notable suffers should have some source). If I did remove any reliable sources in the process I apologise but the number of unreliable sources in the recent edit makes me think Wikipedia:BOLD,_revert,_discuss_cycle may represent the best process for vetting this changes. JamesStewart7 (talk) —Preceding comment was added at 10:44, 7 January 2008 (UTC)

Actually the line you reference above was sourced to a conference paper cited, which contains the wording after a full description of factors "Given Behan does not mention controlling for these factors the improvement------is an amazing result" and therefore is not OR. Well we know your POV concerning RS sources but I think that needs to be tested on a case by case basis if need be in an independent arena such as the Reliable source noticeboard. For instance all Pubmed papers are peer reveiwed, despite your desire for further qualification. Your item regarding words in a title is undefendable except as POV. Hypotheses are worth more than their conclusions, given that they often contain very detailed reveiws of literature. "For this article, where there are reviews and controlled studies available, case studies do not warrant inclusion at all." On this basis then where there is no other source for the identical study then case studies are valid, and therefore included. As you have now reinstated a previous edit that contains conference papers, Neuro endocrinology letters, your summary is a double standard and therefore not relevant commentary. In veiw of these comments I have posted a new revised edit leaving most of your reinstated edit intact. Jagra (talk) 01:43, 8 January 2008 (UTC)

A hypothesis paper may be a reliable source if and only if it is clearly stated in the article that it is a hypothesis only. There are also huge WP:WEIGHT issues with including hypotheses. If we include one hypotheses WP:WEIGHT dictates that we need to include all other hypotheses in proportion to the prominence of each in RS. Hypotheses are not statements of fact yet in the article they have been presented as such. Letters are similarly speculative. There is a huge problem with presenting clearly speculative statements as facts. What edit did I reinstate that contained conference paeprs and neuro endocrinology letters? Is this the original edit of mine which was reverted so that we could discuss the changes and consider removing them again? Because if so that is not a reason to keep them. There is just no way the results of an uncontrolled case study are reliable. Giving them the same attention as a double blind placebo controlled study is a clear cut violation of WP:WEIGHT. The results are not nearly as reliable so why are they getting as much attention as more reliable studies? Also why have conference proceedings been referenced as journal articles. JamesStewart7 (talk) 04:43, 8 January 2008 (UTC)

One more thing. Can we stop including foreing language articles. They are not verifiable unlesss someone can point to a full translation. JamesStewart7 (talk) 04:46, 8 January 2008 (UTC)

I removed this "The findings in CFS of increased lipid peroxidation, oxidative stress and associated lowered antioxidant status, is indicative of hyper responsiveness in CFS, and likely altered immune responses" as a whole bunch of research had been synthesized in a way that constitutes OR. I saw no referenced study directly associated lipid peroxidation, oxidative stress or lowered antioxidant status, with hyper responsiveness in CFS.

This violates WP:WEIGHT. "A third study in 2004 supplemented only with omega-3 essential fatty acid, patients were treated solely with a high-eicosapentaenoic acid-containing essential fatty acid supplement. All showed improvement in their symptomatology within eight to 12 weeks, which it was said may be associated with structual brain changes, seen in a separate case study. There are also reports of an open case series study of 55 consecutive long term incapacitated primary care patients using combined dietary and other EFA modulations instead of supplementation." You don't give uncontrolled and case studies the same weight as double blind controlled studies. They are a far less RS and deserve far less of am ention.

So what are the placebo controls for the supposedly methodologically superior CBT and GET studies, and how do you double blind them? Bricker (talk) 11:42, 8 January 2008 (UTC)

Others have already answered this. You use double-blind studies when you are giving drug treatments. It is easy to construct a placebo for a drug. It is also simple to create a no-treatment control group for a psychological intervention so this is also a requirement. It is extremely difficult (impossible?) to create an appropriate placebo for a psychotherapy which is why it is generally consdiered best practice to compare different types of therapies. For example comparing CBT to GET is reasonable and some stuides have looked at this. As Sciencewatcher pointed out pyschological treatments basically are a type of placebo. A placebo is a psychological intervention, so is psychotherapy. However, this doesn't excuse using uncontrolled studies. The CBT and GET studies are methodologically superior because they actually had control groups and random allocation (maybe non-random patient selection in some studies but they allocate to control/treatment in a random fashion). So yes, an Observational study is not as well controlled as a double blind, placebo controlled trial but no, it is not equal to a case study or an uncontrolled study. There is also no reason to preference an uncontrolled study or case study when RCTs are available as is the case here. If the results of a case study and a RCT trial conflict you go with the results of the RCT every time. The rule is always to preference the highest standard of evidence available. JamesStewart7 (talk) 12:51, 8 January 2008 (UTC)

This "The mixed results of earlier trials was explained as these were also static dose regimes and did mention attempts to control for other intake or lifestyle factors that can effect essential fatty acid utilization." is one of about 50 ways to explain it. It also places the results of the uncontrolled studies above the controlled studies. Neither is it adequately sourced.

This "Amounts of different fats in the diet, amount of protein, alcohol, zinc and magnesium status, exercise amounts, level of conditioning, stress and other infections may also affect the results." is sourced with conference proceedings. In conference proceedings, opinions are presented, not facts. This statment presents facts.

If you want to present facts you need to use scientific studies otherwise you must follow WP:NPOV policy and use the "So and so states..." format. Furthermore you must overcome weight issues and present the other side of the argument (I'll add the reviews from RS in a moment.) JamesStewart7 (talk) 05:12, 8 January 2008 (UTC)

An important issue that is not being taken into consideration is the heterogeneity of the patient selection for the studies. Not all reviews and studies can be properly compared because of it. Case definitions and diagnostic criteria should be described. Also, some of the citations are getting shoddy, please put some work into those too. Thanks Ward20 (talk) 10:05, 8 January 2008 (UTC)

I would agree that we should add the diagnostic critera used in each study referenced. Such types of descriptive statements are reasonable but we should probably avoid judging which is better/more current/reliable etc unless a reviwer stated it as such. The BMJ review did note that differing diagnostic criteria's were used in the EFA study and this may have effected the results so it seems that reviewers are aware of these issues. Maybe something like this, http://www.ncbi.nlm.nih.gov/pubmed/12629919?dopt=Abstract it] can be placed in the diagnostic section as a general barrier to constructing effective treatments. I agree that it does seem like a problem but I think placing it in any particular treatment section may be misleading as it effects the development of all treatments. JamesStewart7 (talk) 12:51, 8 January 2008 (UTC)

No it was not your first revert, your previous revert reinstated the said citations. Have you checked to if and how other hypotheses and conference papers are cited in this Article and if so why have you not removed them or altered weight? When there is even handed treatment I will consider your comments as NPOV. I am not convinced however that conference papers are not RS, what can you cite to show that is not POV? Until then I have reinstated them. The ‘mixed results’ statement was sourced to a conference paper, as was the ‘Amounts of different fats in the diet’ statement which was sourced to the same conference paper, because this source links the statement with CFS specifically. Also it was sourced to an hypothesis paper and a third ref. PMID 2001722. I have removed the Hypothesis paper until I have checked whether the relevant words are a conclusion or in a literature summary section. I will also bring worth the words in the remaining paper. Have you checked the full n-3 paper to see what type of study it actually was before reverting on a claimed case series study? This was included as it is the only study using solely n-3 EFA’s as disatinct from the other cited trials. The actual case study was included because it the only study that has examined Cerebral magnetic resonance scanning in association with EFA supplementation, and as was said "For this article, where there are reviews and controlled studies available, case studies do not warrant inclusion at all." “On this basis then where there is no other source for the identical study then case studies are valid, and therefore included “. These are placed below the better studies, and even if all are as described, undue weight does not mean no mention at all. So I have altered the wording. The supposed synthesis comment will be dealt with separately. There is a serious mistake I believe in the current wording that you should check as sunflower oil is not a placebo in EFA trials as it contains high amounts of n-6 EFAs and is a treatment in itself. If Warren did actually use sunflower oil as placebo, little wonder there were no differences with treatment? or is it the case that the reveiw has screwed up badly? In which case it demonstrates why not to quote reveiws, and this one should not be quoted at all. So I have removed these words. There is no specific currently accepted diagnostic criteria, they are all accepted as relevant, so have removed this POV statement as well. Without better citing I cannot verify the Reid/Chadler review, perhaps you can provide a link? Jagra (talk) 07:17, 9 January 2008 (UTC)

Well apparently I'm not the only one who thinks the sources have reliability problems as someone else reverted it this time. There are other studies on EFA supplementation and yes I did look at the study and there was no control group. Just because the study is slightly different does not mean that it warrants inclusion. Unless someone is specifically attempting a replication, two studies are rarely the same. Please note that NPOV policy does not state that we should give all views equal consideration. It states that we should give more weight to the POV with more reliable sources and we should allocate POV in proportion to the reliability of those sources. The wording about the sunflower statement was the best paraphrase I could come up with from the BMJ journal. If anyone thinks the can more reliably capture what the BMJ review authors stated then go ahead but bear in mind the goal is to preserve the author's comments, not insert our own. The authors also commented on the diagnostic criteria. I also believed I wrote the comments in the most NPOV form possible, using the phrase "The researchers state..." before introducing their opinion.

Also can you please point to exactly what unreliable citations I have reinstated so that I may correct this. It is very hard to check myself because I am unsure which edits you are referring to. JamesStewart7 (talk) 09:24, 9 January 2008 (UTC)

Bricker, CBT can and should be compared to other psychological therapies to help determine the active or differentiating ingredient. Comparisons to counselling has yielded interesting results. I'll post the rest of my reply in a new section titled "CBT vs counselling". - Tekaphor (talk) 06:30, 10 January 2008 (UTC)

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.