RACP Guidelines for CFS

Jim Oakley

1. General Overview

The Guidelines have been refocused somewhat to be of more practical use to doctors. The chapter on 'CFS in Adolescents' is an important and welcome addition, and should be of significant help to GPs and specialists in managing young patients. The chapter on 'Managing CFS' is more mature, comprehensive and informative than in the previous draft. It is a good approach to recommend a variety of conceptual frameworks for illness management, so that different patient preferences and medical skills can be accommodated. The emphasis on tailoring a rehabilitation/behavioural/cognitive program to take into account an individual's level of disability, personal preferences and response to treatment is pleasing, and will help to promote better outcomes and fewer adverse experiences for patients.

Yes, it is true that since the first draft of the guidelines more evidence has been published supporting the use of cognitive behavioural therapy (CBT) and graded aerobic exercise (GAE) in people with CFS, but the available data have limitations, as does the discussion of this subject in the new draft (see comments below).

While some of the 'rough edges' in the first draft have been smoothed over, a range of other issues has not been addressed, and discussion of these constitutes the bulk of my present submission.

I can't check the tables since they were not included in the version of the new draft sent out for comment, but in answer to Rob Loblay's query I do think that Tables 3.1, 3.2, 4.1 and 4.2 in the first draft are important and should be retained and updated, and I also support the deletion of the FAQ box. The old Chapter 5 has disappeared, but the tables in this chapter were useful, and I would like to see updated versions of these in the final version of the guidelines.

2. CBT/Graded Aerobic Exercise

At the end of the chapter on 'Managing CFS' it is stated that:

"Doctors are encouraged to discuss frankly issues about evidence – including what is known and what is not kn– to ensure that patients are able to make informed decisions about treatment."

If I go to a surgeon and he recommends that I have an operation, he will be able to tell me the success rate for that particular operation and what the potential complications may be. If a doctor recommends to a person with CFS that he/she should undertake a program of rehabilitation involving CBT and/or GAE, the first questions likely to be asked by the patient will relate to the nature of the program, the likely outcome, and potential problems or drawbacks.

Undertaking a CBT/GAE program will be a serious, multi-month endeavour for someone with CFS, and it is important that the doctor and patient embark on such a program with realistic expectations and a common understanding of what possible outcomes may eventuate.

Unfortunately, at present the guidelines do not contain sufficient information about published CBT/GAE studies to allow doctors to discuss issues about evidence, or to allow patients to make informed decisions about treatment options. The guidelines must be open and honest with doctors and patients. It is essential that the guidelines contain a table with details of published controlled CBT/GAE trials including the percentage of participants who improved with treatment, the percentage who remained the same, the percentage who deteriorated and the percentage who dropped out (where this information is available), along with details of the nature and duration of the treatment.

The guidelines should also clearly acknowledge limitations in the available data. For example, published GAE/CBT studies have only involved ambulatory patients, not those bedridden or wheelchair bound. Similarly, only adults have been included in these studies, not adolescents or children. Thus at least two important groups exist for whom there is no published data about the efficacy of GAE/CBT. The guidelines must acknowledge this if doctors and patients are to make "informed decisions" about treatment.

With CBT, the evidence is that better results are obtained when the therapy is administered by therapists who are highly trained in the relevant techniques. For example, the very large multicentre CBT trial done in the Netherlands employed therapists only recently trained in the technique for CFS, and the authors of the study noted that their results "showed a lower proportion of patients with improvement than CBT trials with a few highly skilled therapists" [Lancet 2001 Mar 17; 357(9259):841-7]. A recent study of CBT conducted in ten general practices in the United Kingdom found that CBT was no more effective than counselling for people with chronic fatigue, and the the same result applied to the subset of people with CFS in the trial [Br J Gen Pract 2001 Jan; 51(462):19-24]. Again, improvement rates in this trial were lower than in those which used highly skilled therapists.

This is important information to convey to doctors and their CFS patients who may be considering undertaking a CBT program, since it is unlikely that GPs will have any expertise in delivering CBT programs in general, let alone to people with CFS in particular. A referral to a specialist CBT therapist may be in the best interests of the patient, since a better outcome is more likely (based on the studies mentioned above). Once again, if patients and their doctors are to make "informed decisions about treatment" then the guidelines must provide more information to readers than they currently do.

In the 'Clinical Overview' it is stated that:

"Graded exercise programmes have been shown to be safe for people with CFS..."

and in the 'Management' box it is stated that:

"Graded aerobic exercise is safe and effective for people with CFS (Level II)".

Sweeping statements like these are simply not accurate, and would be recognised as inappropriate if the actual study results were presented in the guidelines for readers to assess. The statements above should be qualified with words like "many", or with actual percentage outcomes derived from studies. For example, in the 'Clinical Overview', the sentence should read:

"Graded exercise programmes have been shown to be safe for many people with CFS..."

or

"Graded exercise programmes have been shown to be safe for up to xx% of people with CFS..."

3. Sleep Management

The thrust of the recommendations about sleep management in the chapter on 'Managing CFS' is fine, and it is hard to argue with the ultimate goal of restoration of the normal sleep-wake cycle and normalisation of circadian rhythm. However, the specific recommendations for sleep duration in the 'Managing CFS' chapter are significantly at odds with the personalised, tailored, feedback-driven approach advocated in the guidelines for CBT/GAE.

It should be noted that in the 'Phenomena associated with CFS' box, it is stated that:

"Sleep disturbance is very common in people with CFS (Level I), but is of uncertain pathophysiological significance"

and in the 'Clinical Overview' it is stated that:

"Clinical experience suggests that sleep interventions in people with CFS may reduce symptoms and improve functional capacity, although direct evidence for this is currently lacking".

Thus there is uncertainty (indeed a lack of evidence) about the role of sleep disturbance in the perpetuation of symptoms and disability in people with CFS. The rigid, highly prescriptive recommendations for sleep duration in the guidelines do not reflect this uncertainty, and leave little room for doctors to accommodate the different levels of health and different personal circumstances of their CFS patients.

Furthermore, hypersomnia is a recognised feature of CFS in some people, particularly those in the early stages of the illness, and in those more severely affected. "The need for excessive sleep" was present in 87% of patients in a survey of 290 adolescents with CFS, according to the 'CFS in Adolescents' chapter. It is very likely that hypersomnia is a direct result of the disease process in CFS, and not an indirect consequence of illness (which is the case with a shift to a late night/late morning sleep cycle).

In particular, two sleep recommendations in the 'Managing CFS' chapter need modification. The first is:

"restrict the night-time sleep period to about eight hours"

and the second is:

"reduce (to less than 30 minutes) or abolish daytime naps".

While these recommendations may be suitable for someone with CFS who has a only a mild sleep disorder, or who is not severely affected by CFS, they are much too restrictive for someone who is very sick and has associated hypersomnia. What is the point of making recommendations that are not attainable by many people (and not even approachable in the case of some)? This will only cause alienation, and a lack of faith in this and other aspects of the sleep management program.

The problems with the above italicised quotes can be easily rectified by the inclusion of qualifying statements indicating that the duration of sleep at night – and of naps during the – may need to be varied according to the patient's level of health, and/or that it may take some time before the patient is able to comply with the recommendations. Slow progress towards a desirable goal is better than none at all, as is acknowledged in the section on CBT/GAE.

Let's face it, much of the world's normal, healthy population would have trouble complying with the sleep recommendations in the guidelines, particularly the multitude who routinely have a siesta during the day, in addition to the significant percentage of people who naturally sleep more than eight hours per night.

By the way, as with the first draft of the guidelines, it is still not clear in the current draft whether sleep apnoea is an alternative primary diagnosis to CFS, or whether it can exist concurrently with CFS. There are several apparently contradictory statements on this matter in the current draft which need to be corrected or clarified.

4. Placebo Effect

In my submission to the Working Group in response to the public release of first draft of the guidelines, I presented a table listing all published controlled trials of treatments for CFS and the response to either placebo (in the case of pharmacological treatments) or 'non-active' treatment (in the case of behavioural treatments) that occurred in these trials. Up to January 1998, there were 15 published pharmacological trials for CFS which included a placebo control arm. Of these, three studies reported placebo response rates above 30% (44%, 42% and 33% respectively), two did not provide numerical data but noted that the placebo response was "statistically significant", one paper did not present placebo response data in the abstract (I listed this paper as 'unsighted' since I only had the abstract), and the remaining nine studies all reported placebo response rates of less than 20%, including four which had a 0% placebo response rate. Some studies with high placebo response rates included large numbers of patients, and/or people with very chronic and disabling forms of CFS. Similarly, some studies with low placebo response rates included large numbers of patients and/or people with very chronic and disabling forms of CFS.

Placebo response data for CFS trials published after January 1998 are presented below, in chronological order:

Treatment

Staphylococcus toxoid vaccine
N (total) 28
Design Parallel
N (placebo) 14 (inferred)
Placebo response No significant improvement except in pain levels
Reference Andersson M et al, Eur J Pain 1998; 2(2):133-142

Treatment

L-glutamine
N (total) 16
Design Parallel
N (placebo) 8 (inferred)
Placebo response Not stated in abstract
Reference Rowbottom D et al, Journal of CFS 1998; 4(2):3-22

Treatment

Fludrocortisone
N (total) 25
Design Crossover
N (placebo) 20
Placebo response 0% (no improvement in any measure)
Reference Peterson PK et al, Arch Int Med 1998 Apr 27; 158(8):908-14

Treatment

Growth Hormone
N (total) 20
Design Parallel
N (placebo) 10 (inferred)
Placebo response Not stated in abstract
Reference Moorkens G et al, Growth Horm IGF Res 1998 Apr; 8 Suppl B:131-3

Treatment

Fluoxetine and graded exercise
N (total) 136
Design Parallel
N (placebo) Not stated in abstract
Placebo response Not stated in abstract
Reference Wearden AJ et al, Br J Psychiatry 1998 Jun; 172:485-90

Treatment

Hydrocortisone
N (total) 70
Design Parallel
N (placebo) 35
Placebo response 54% showed some improvement on Wellness score, although only 29% had an improvement of 5 points or more on the Wellness score (compared with 53% in the treatment group).
Reference McKenzie R et al, JAMA 1998 Sep 23-30; 280(12):1061-6

Treatment

Essential Fatty Acids
N (total) 50
Design Parallel
N (placebo) 25 (inferred)
Placebo response No significant improvement in symptoms
Reference Warren G et al, Acta Neurol Scand 1999 Feb; 99(2):112-6

Treatment

Oral NADH
N (total) 26
Design Crossover
N (placebo) 26
Placebo response 8% "responded favourably"
Reference Forsyth LM et al, Ann Allergy Asthma Immunol 1999 Feb; 82(2):185-91

Treatment

Hydrocortisone
N (total) 32
Design Crossover
N (placebo) 32
Placebo response 9% had significant reductions in fatigue
Reference Cleare AJ et al, Lancet 1999 Feb 6; 353(9151):455-8

Treatment

Moclobemide
N (total) 90
Design Parallel
N (placebo) 43
Placebo response 33% improved
Reference Hickie IB et al, J Clin Psychiatry 2000 Sep; 61(9):643-8

Treatment

Fludrocortisone
N (total) 100
Design Parallel
N (placebo) 50
Placebo response 10% experienced at least a 15 point improvement in their wellness scores (which was the predetermined main outcome measure)
Reference Rowe PC et al, JAMA 2001 Jan 3; 285(1):52-9

Of these eleven additional pharmacological trials, two reported a placebo resonse rate greater than 30%, six reported a placebo response rate of 10% or lower (including two studies which reported a 0% placebo response), and for three studies there was no placebo response data presented in the abstract.

The numbers of placebo recipients were 35 and 43 respectively for the two studies with placebo response rates greater than 30%; for the six studies reporting placebo responses of 10% or lower, placebo group sizes were 14, 20, 25, 26, 32 and 50 respectively.

In summary, of the 26 published double-blind placebo controlled studies of pharmacological treatments for CFS, 15 have reported placebo response rates of less than 20% (6 of which had a 0% placebo response rate), 5 have reported placebo response rates greater than 30%, 2 have reported a "statistically significant" placebo response, and 4 did not report placebo response data in the abstract (I did not have access to the full papers in these cases).

In the chapter 'Managing CFS' under the heading 'Unproven therapies in CFS', the first paragraph contains the following sentences:

"Given the variable clinical course of CFS, the likelihood of spontaneous improvement, and the high rate of non-specific (placebo) responses, properly controlled clinical trials are essential for the evaluation of all proposed new treatments.....It should be borne in mind that in most controlled trials at least 30-50% of people with CFS demonstrate improvement in the non-specific (placebo) treatment arm (Wilson et al. 1994a; Hickie et al. 1995b)".

These statements about high placebo response rates are clearly wrong; they refer only to a minority of published studies, and they must be removed from the guidelines. This is a matter of scientific integrity, as well as fair play. Furthermore, in my previous submission I pointed out the gross inadequacy of data about placebo response rates in the supporting references (Wilson et al 1994a, and Hickie et al 1995b).

Also in the chapter 'Managing CFS', under the heading 'Principles of management' is the statement:

"The significant non-specific (placebo) response rate in controlled treatment trials for people with very chronic and disabling forms of CFS is likely to be a reflection of these essential components of good clinical practice".

This is another misleading statement. Please note that not all significant placebo response rates occured in studies of people with very chronic and disabling forms of CFS, and in published studies it is certainly not true that people with very chronic and disabling forms of CFS usually report high placebo response rates. Several trials with long term, severely affected sufferers have shown very low placebo response rates [eg. Peterson et al. 1998; Strayer et al. 1994; Vercoulen et al 1996; and others].

The above statement in the guidelines should be modified so that it doesn't imply that most or all treatment trials with people with very chronic and disabling forms of CFS have high placebo response rates. An alternative wording is:

"The significant non-specific (placebo) response rate reported in a minority of controlled treatment trials for people with CFS is likely to be a reflection of these essential components of good clinical practice".

5. Predictors/Associations of Outcome

The discussion in the guidelines about the natural history of CFS relies heavily on the review of Joyce et al 1997. Unfortunately, this review is littered with errors and is consequently unreliable. In fact, one of the submissions in response to the first draft of the guidelines was from an American psychologist who was concerned about the accuracy of aspects of natural history chapter, including the reliance on the review of Joyce et al. He had contacted the senior author of the paper of Joyce et al, Professor Simon Wessely, and had discussed the errors with him. Professor Wessely said words to the effect that it was regrettable that so many errors in the paper had found their way through the peer review process and into publication.

Many research groups have invested much time and effort to collect natural history data about their CFS patients, and have produced comprehensive and valuable reports. Their efforts should not be diminished or distorted by references to an unreliable review. The Working Group must examine each and every original paper on the natural history of CFS and then draw conclusions, and should make no mention of the review of Joyce et al.

In my submission in response to the first draft of the guidelines, I presented data from the five large longitudinal studies published up to that time which had investigated predictors of outcome for people with CFS. I excluded two other studies from the analysis due to lack of use of a published case definition for CFS, and/or very small sample sizes.

The factors that were consistently associated with poor outcome, for which there was little or no conflicting evidence, were as follows:

The factors that were consistently found to be not associated with poor outcome, and for which there was little or no conflicting evidence, were:

Since the publication of the first draft, a further six longitudinal studies have been published which examine predictors and associations of outcome. These are:

  1. Tiersky LA et al, Appl Neuropsychol 2001; 8(1):41-50
  2. Hartz AJ et al, Arch Fam Med 1999 Nov-Dec; 8(6):495-501 (idiopathic chronic fatigue)
  3. Pheley AM et al, Minn Med 1999 Nov; 82(11): 52-6
  4. Hill NF et al, Arch Phys Med Rehabil 1999 Sep; 80(9): 1090-4
  5. Russo J et al, J Psychsom Res 1998 Jul; 45(1 Spec No): 67-76
  6. Miro O et al, Med Clin (Barc) 1997 Apr 19; 108(15): 561-5

In addition, two recent treatment trials have looked at prognostic factors for outcome after treatment:

  1. Deale A et al, J Psychosom Res 1998 Jul; 45(1 Spec No.): 77-83
  2. Prins JB et al, Lancet 2001 Mar 17; 357(9259): 841-7

In general, the new studies provide additional evidence to support the predictors and associations of outcome that I have listed above. Psychiatric disorder at follow-up was associated with poor outcome in the studies of Russo et al 1998, and Hartz et al 1999. High levels of fatigue or functional impairment at study entry were found to predict poor outcome in the study of Pheley et al 1999. A sense of control over symptoms at entry was found to predict better outcome in the treatment trial of Prins et al 2001.

Contrary to earlier studies, age was found to be a significant predictor of outcome in the study of Tiersky et al 2001, but very few other studies support this finding in adult patients. Gender has not predicted outcome in any published study. Emotional disorder at study entry did predict outcome in the study of Tiersky et al 2001, but not in the study of Hill et al 1999.

In the chapter 'What is CFS?' under the heading 'What is the outcome of fatigue states?' is the statement:

"Factors associated with a poorer outcome include older age, concurrent psychiatric disorder, and the person's belief that the illness is purely physical in origin".

and in the 'Natural history' box is the sentence:

"Older age, psychological disorder, and belief in a purely physical cause of CFS are associated with poorer outcomes (Level II)".

It is clearly not reasonable to maintain that older age is associated with poorer outcome (in adults) since the overwhelming weight of evidence is against this statement (see comments in my previous submission). Psychological disorder at follow-up is associated with poor outcome. As I discussed in my previous submission, there is no clear evidence that a person's belief that the illness is purely physical in origin is associated with poor outcome, since two studies have found this association exists (one only measured this belief at follow-up, not at baseline), and two studies have not. More recently, the treatment trial of Deale et al 1998 found that physical illness attributions were not associated with poor outcome in either the treatment or control groups.

The two statements in the guidelines quoted in italics above should be modified to list the following three factors which are associated with poorer outcome:

It is also stated in the same paragraph of the guidelines that:

"Outcome has not been found to be associated with gender, marital status or life stress events..."

Please note that two studies [Hartz et al 1999 and Miro et al 1997] did find that being married was associated with better outcome.

In the chapter 'Managing CFS' under the heading 'The role of rehabilitation, behavioural and/or cognitive treatment approaches' is the statement:

"Given that simplistic illness attributions are associated with poor outcome...".

It is not reasonable to say this, given that the weight of evidence is somewhat in favour of the converse being true. Please change or delete this statement.

In the chapter 'Social & Legal Issues' under the heading 'What are the disadvantages of a diagnosis of CFS?' is the statement:

"Thus, when the prognostic features are favourable (i.e. younger age group; less severe symptoms; shorter duration of illness; and onset with a specific infection, eg. glandular fever)..."

I am unaware of any published study which has found that onset with a specific infection is a favourable prognostic feature. On the contrary, in the study of Hill et al 1999, mode of onset did not predict illness outcome. In the study of Reyes et al 1999 [Journal of CFS 1999; 5(1): 17-27], both sudden onset and gradual onset cases had similar probabilities of recovery during the first 15 years of illness. Please remove the words "and onset with a specific infection, eg. glandular fever" from this sentence in the guidelines.

6. Other Issues

Idiopathic Chronic Fatigue

I remain unconvinced that the guidelines adequately deal with the issue of idiopathic chronic fatigue, ie. clinically investigated, unexplained fatigue of six months or more which does not meet all the criteria for CFS. If Box 1.1 from the first draft of the guidelines is retained, then 'chronic fatigue' will refer to both uninvestigated fatigue of more than six months duration, as well as to idiopathic chronic fatigue. For people who do have idiopathic chronic fatigue, no particular label for their condition is recommended in the text of the guidelines; rather, a name for their illness ('chronic fatigue') has to be inferred from Box 1.1 or Box 2.2 (in my previous submission I pointed out inconsistencies between these two boxes).

In the 'Social & Legal Issues' chapter, under the heading 'Medicolegal issues', is the following statement:

"In verifying a diagnosis of CFS, the current international diagnostic criteria should be applied, including documentation of the characteristic symptoms, the lack of abnormalities on physical examination and results of the recommended laboratory investigations".

What the status of people with idiopathic chronic fatigue will be in a medicolegal setting is uncertain, because the guidelines do not mention the term 'idiopathic chronic fatigue', and are therefore inconsistent with the paper by Fukuda et al 1994. Furthermore, the phrase 'chronic fatigue' has two meanings in the guidelines (as pointed out above), but only one meaning in the paper of Fukuda et al.

People with idiopathic chronic fatigue are not necessarily any less sick than people who meet the narrow research criteria for CFS. In fact, it is likely that a large number of people in the community who have been given the diagnosis of CFS actually have idiopathic chronic fatigue, because they don't meet all the research criteria for CFS. The guidelines should provide a clear and unambiguous diagnostic label for this group of people that is consistent with the definition in the paper by Fukuda et al 1994.

Fludrocortisone Studies

Two double-blind, placebo-controlled trials of fludrocortisone in people with CFS have been published in major medical journals (JAMA and Arch Intern Med). These trials and their results should be discussed in the 'Managing CFS' chapter, as has been done for the two trials of hydrocortisone and the four trials of immunoglobulin.

Food Intolerance

The guidelines suggest the judicious use of a range of unproven pharmacological treatments to treat different symptoms in people with CFS (see 'Symtomatic drug treatment' in the 'Clinical Overview'). In fact more evidence exists to justify the treatment of food intolerance in CFS than exists to justify the use of agents such as amitryptyline, sertraline, paroxetine or nefazodone. The not uncommon problem of food intolerance in CFS should be mentioned in the guidelines, perhaps with a recommendation for specialist assessment and treatment when it is indicated.

Urinalysis for Paediatric Patients

Just a query: in the 'Clinical Overview' and in the 'Laboratory investigation' box, the recommended laboratory tests for adult CFS patients include urinalysis for protein, blood and glucose. However, in the chapter 'CFS in Adolescents' the list of recommended laboratory tests for paediatric patients does not include urinalysis. Is this an oversight or a deliberate omission? Is urinalysis relevant for excluding other diseases in children and teenagers?

Mood Alteration in CFS Diagnostic Criteria

In the chapter 'What is Chronic Fatigue Syndrome' under the heading 'Does chronic fatigue overlap with other illnesses' is the statement:

"The high rate of comorbidity is not surprising as diagnostic criteria for both CFS and major depression (DSM-IV; ICD-10) include fatigue, sleep disturbance, cognitive impairment and mood alteration".

Actually, the diagnostic criteria for CFS used in the guidelines (Fukuda et al 1994; see Box 1) do not include any reference to mood alteration. Please correct this sentence in the guidelines.

Cost Update

In the chapter 'What is Chronic Fatigue Syndrome' under the heading 'What is the cost of CFS to the community?', it would be more helpful and relevant if costs were converted to 2001 dollars instead of 1997 dollars.

Inappropriate Reference – 1

Yet again I wish to point out the inappropriateness of the reference to the paper by Sharpe et al 1992 in two particular sentences in the guidelines.

In the chapter 'What is chronic fatigue syndrome?' under the heading 'What is the outcome of fatigue states?' it is stated:

"Such patient samples are biased towards chronic illness and limited patterns of recovery. Patient reports drawn from self-help group populations show similar biases (Sharpe et al. 1992)."

In the chapter 'Social & Legal Issues' under the heading 'The role of patient support groups' is the statement:

"Inevitably, they tend to attract those with the most prolonged illnesses (Sharpe et al 1992)Š."

Please note that Sharpe et al 1992 found a correlation between functional impairment and membership of a patient organisation. Their study found no correlation between length of illness and membership of a patient organisation. I request that all Working Group members read this paper if there remains any doubt about what correlations were found. The references to Sharpe et al 1992 in the italicised quotes above should be removed from the guidelines and replaced with an appropriate reference (perhaps from one of the other 'patient reports drawn from self-help group populations'). The statements should be deleted if they can't be supported with published evidence.

Inappropriate Reference – 2

In the chapter 'What is Chronic Fatigue Syndrome' under the heading 'What is known about the pathophysiology of CFS?', a list of "leading hypotheses" about the pathophysiological basis of CFS is presented. The fifth hypothesis in the list is:

"a psychologically determined response to infection or other stimuli occurring in 'vulnerable' individuals (Imboden et al. 1961; Abbey 1993)".

In my previous submission I detailed the complete inadequacy of the two references used to support this statement. They must be removed, and replaced with references to valid and relevant studies. If no studies have tested this hypothesis, or if it has been disproved, it cannot be regarded as a "leading hypothesis" after more than 15 years of research into CFS.

Biased Use of Evidence

In the chapter 'What is Chronic Fatigue Syndrome' under the heading 'What is the outcome of fatigue states' is the following sentence:

"In a study of the relationship of non-specific viral illness and the development of prolonged fatigue in general practice, the person's view of the illness and the doctor's behaviour, rather than the viral infection, were predictive of the development of prolonged fatigue (Cope et al. 1994)".

This is a highly selective use of the available evidence. Another larger study (Wessely at el 1995), which followed people from before the onset of a viral illness until six months after their initial presentation to a general practitioner, reported different results. Wessely et al 1995 found that the person's view of the illness (ie. belief that fatigue was due to a physical cause) was not a significant predictor of fatigue at six months' follow up. This is in direct contrast to the findings of Cope et al 1994.

Why have the results of one study been selectively quoted, while the results of the other – which presents a different v– have been ignored? This biased approach does not lend credibility to the guidelines, and distorts the impression readers will obtain about factors which influence the development of prolonged fatigue.

Unhelpful Use of Evidence

In the chapter 'What is chronic fatigue syndrome?' under the heading 'Does chronic fatigue overlap with other illnesses?' appears the statement:

"Up to two-thirds of adults with CFS have either a prior, or concurrent, diagnosis of major depression, as do people with fibromyalgia and irritable bowel syndrome. By comparison, the lifetime rate of comparable depressive disorders in the general community is 15%-25%."

As I pointed out in my previous submission, it would be more helpful for doctors and other readers if the figures for prior and concurrent depression in people with CFS were quoted separately. It would be useful for GPs to have an estimate of the likelihood of concurrent depression in their CFS patients, but lumping prior depression in with concurrent depression simply confuses the issue and makes the incidence of depression in CFS appear higher than it actually is. Also, to be consistent with the presentation of data for community lifetime depression, an upper and lower range should be quoted for the incidence of prior and concurrent depression in people with CFS.

Readers are informed in the chapter 'What is chronic fatigue syndrome' under the heading 'What is the outcome of fatigue states?' that:

"The long-term outcome of CFS has been evaluated mostly in people treated within tertiary referral settings. Such patient samples are biased towards chronic illness and limited patterns of recovery".

Rates of depression in CFS have also been evaluated mostly in people treated within tertiary referral settings. To be fair and consistent, readers should be informed of this, since the frequency of depression may be higher in patient samples biased towards chronic illness and limited patterns of recovery.

Miscellaneous

In the 'Epidemiology' box, the figures for the prevalence of CFS in the community are stated to be 0.2% - 0.5%. In order to be consistent with the text (and the published studies), the correct figures of 0.2% - 0.7% should be used.

Return to Top

Alison Hunter Memorial Foundation
PO Box 6132 North Sydney 2059 Australia
Phone/Fax +61 2 9958 6285

Home
About Us
About ME/CFS
Severity
Advocacy
Research
Guidelines
Conferences
Medical Politics
Media
Archives
Links
Donations