ME/CFIDS/CFS/FM Unhelpful Counsel
Continued from Feb 2 Unhelpful Counsel

3. Limitations of the report

The positive aspects of the report listed in Section 2 concern the recognition of ME and the need for illness management in a variety of forms. However, the report has limitations. Some, such as the constrained remit, are obvious, but others are apparent only to those au fait with the issues, whether the research evidence or the deliberations of the Working Party.

3.1 Constrained remit
The report describes its own remit clearly:

"To review management and practice in the field of CFS/ME with the aim of providing best practice guidance for professionals, patients, and carers, to improve the quality of care and treatment for people with CFS/ME, in particular to: develop good clinical practice guidance on the healthcare management of CFS/ME for NHS professionals, using best available evidence; make recommendations for further research into the care and treatment of people with CFS/ME; identify areas which might require further work and make recommendations to CMO." (1.1)

Given the controversy surrounding the illness, this remit seems primarily designed to contain and manage the clinical problem. By concentrating principally on management and `guidance', the report has ensured that the focus is on containment and coping, rather than on addressing the clinical conundrum of causation. Though the cause of CFS/ME has yet to be elucidated, it is important to consider the various possibilities at length within the context of management options. The narrow focus also neatly sidesteps the problem of the preference of some clinicians for the umbrella term `CFS', obscuring specific diagnosis and possibly, in the long run, influencing management and practice for the worse.

The result is that the Working Group has taken three years to uncover the obvious - that, for a variety of reasons, the available `management strategies' are cognitive behavioural therapy and graded exercise therapy, both with a very weak and rudimentary evidence base (Whiting et al, 2001), and `pacing', which is little more than a commonsense approach to physiological limitation. Welcome though recognition of the illness is, we should not forget that for patients and carers little if anything has changed, or probably will change, as a consequence of developing `best practice guidance'.

3.2 Unbalanced composition of the Working Group

The final Working Group consisted of Professor Allen Hutchinson (Chair), Dept. of Public Health, Sheffield; Professor Anthony J Pinching (Deputy Chair), Dept. of Human Science and Medical Ethics, St Bartholomew's Hospital, London; Dr Tim Chambers (Chair of the Children's sub-group), Southmead Hospital, Bristol; and three groups - the Key Group (responsible for surveying the evidence, developing the main report, and agreeing the final recommendations to the CMO), the Children's and Young People's Group, and the Reference Group (with an ad hoc advisory/consultative role). The Report also concedes the input of un-named others not included in the above groups. The breakdown of the Working Group by professional interest is shown in Table 1 below. Due to a misprint in the final CMO report (page 72), the members of the Children's Group are designated as "Key Group Observers", though the existence of the Children's Group is mentioned at several points in the text. Table 1, however, shows the composition of both groups.

Table 1. Composition of the Working Group - excluding the Chair and Deputy Chair - by professional interest at September 2001

	Key group	Children's group
Patient representatives	3 (2*)	1
Carers	1	2
Representatives of ME associations	3	4
GPs	1	0
Psychiatrists/psychologists	3 (3*)	1 (1*)
Public health specialists	2 (1*)	0
Paediatricians	0	3
Nurses	0	1
Council/NHS	0	1

(*In parentheses are the numbers of this professional interest group who refused to endorse the final Working Group report in January 2002. Four of these were professionals with an affiliation to psychiatry/psychology, one was a consultant in Public Health Medicine, and two were patient representatives.)

Given the range of clinical signs and symptoms exhibited by CFS/ME patients, the volume of published research evidence on pathophysiological mechanisms, and a large body of professional opinion supporting a `biomedical' model of the illness, it seems negligent that three of the clinicians on the Key Group should have been psychiatrists/psychologists and that four should have been co-authors of scientific papers supporting the use of biopsychosocial interventions for CFS/ME. Interestingly, the Deputy Chair of the Working Group in a recent paper (Pinching, 2000) advocated the use of the management strategies - cognitive behavioural therapy and graded exercise - finally identified as the therapeutic interventions of choice by the Working Group. The composition of the Children's Group was less skewed towards the psychiatric or the psychological, both of which are generally recognised to be far less appropriate models in children.

In total, six members (46%) of the Key Group refused to endorse the final report, a remarkable attrition rate for a Working Group reporting to the Chief Medical Officer of England. The four professional resigners from the Key Group argued that the report paid too little attention to biopsychosocial aspects (Clark et al, 2002). In Hospital Doctor, 17th Jan 2002, one of them, Dr Alison Round, was reported as saying that the report neglected the "biopsychosocial" aspects of the illness. Another, Dr Peter White, was reported to say, "All the evidence taken together suggests that the condition is biopsychosocial - both physical and mental factors are involved." This kerfuffle has not been universally welcomed: in a recent debate on CFS/ME in the House of Lords, Lord Clement-Jones said, "Some recent articles written by doctors in the wake of the report are absolutely disgraceful and ignorant. I feel strongly about some of those reactive reports" (Hansard, 2002). Patients and carers can only speculate on the kind of report that might have emerged (and the different emphases that might have been placed on psychological strategies) if these professionals had `resigned' at the beginning of the process rather than at the end. As a patient said wistfully, "After all that... it's like cuckoos leaving their trademarks but not their signatures."

3.3 Problem of diagnosis and use of the composite term CFS/ME

Terminology is the `hot' issue in ME and CFS: it energises the debate between patients and healthcare professionals, particularly psychiatrists. It also impacts on patient management and clinical practice since the results of clinical trials are determined by entrance criteria used to recruit patients to them.
The issue can be simply put. The original case description of the illness, `ME' (Acheson, 1959; Dowsett et al, 1990) described a condition, commonly of infectious onset, characterised by:
Exercise-induced fatigue precipitated by trivial exertion (physical or mental). Neurological disturbance, especially of cognitive, autonomic, and sensory functions. This could include impairment of short-term memory and loss of powers of concentration, usually coupled with emotional lability, nominal dysphasia, disturbed sleep patterns, dysequilibrium and/or tinnitus. An extended and relapsing course with fluctuation of symptoms, usually precipitated by either physical or mental exercise; typically, the symptoms vary capriciously from hour-to-hour and day-to-day with varying involvement of the cardiac, gastro-intestinal, and lymphoid systems.

Since the late 1980s, however, the medical profession has been urged by some of its members to adopt the term Chronic Fatigue Syndrome (CFS), a more wide-ranging diagnostic category which includes patients whose dominant symptom is medically unexplained, on-going, or chronic fatigue (in conjunction with several other physical or psychological symptoms) who would not necessarily fulfil the criteria for ME.

There are now several definitions of CFS. In the USA, the 1994 CDC case-definition of CFS is currently utilised (Fukuda et al, 1994), supplanting its predecessor, the 1988 CDC criteria, and has similarities with - but is not identical to - the classical description of ME. However, in the UK, a frequently-used case definition is the `Oxford criteria' which includes patients with no physical signs and inadvertently selects subgroups of patients with high levels of psychiatric diagnoses (Katon & Russo, 1992; Freiberg, 1999). These definitions have been used to recruit to randomised clinical trials, including some of the trials of `psychological' interventions, cognitive behavioural therapy, and graded exercise therapy, which form the basis of the management strategies uncovered by the Working Group report. Since the adoption of a particular case-definition of CFS will greatly influence the outcome of particular studies, it is perhaps no surprise that psychiatric research groups researching biopsychosocial strategies in these patients should find some encouraging results. However, as many patients and carers in CFS/ME support groups in the UK invariably point out:

Fatigue is not their primary problem: musculoskeletal pain and post-exertional myalgia along with other physical signs are far more prominent, corresponding more closely to the classical definition of ME. The World Health Organisation International Classification of Diseases (ICD) has, since 1969, classified ME separately as a neurological problem (ICD 10 93.3), with `CFS' incorporated into the current ICD as a sometime synonym for ME. The chronic fatigue states per se are listed under mental and behavioural disorders (F 48.0), a category which specifically excludes ME/PVFS/CFS.

For these reasons, many CFS/ME patients - particularly the most severely affected - resent being provided with non-curative coping strategies, such as cognitive behavioural therapy, by healthcare workers who have no interest in their particular symptom complex. In this, they are supported by a growing number of experts who consider that there is a strong, perhaps overwhelming, case for unpacking the term `CFS' and reclassifying and renaming in accordance with more specific clinical criteria (De Becker et al, 2001; Tan et al, 2002), such as the criteria for ME described above. The report alludes to the problem:
"The issue of subgroups or discrete entities within CFS/ME was the subject of much debate by the Working Group. We are conscious that some sectors strongly hold the view that the term ME defines a subgroup within CFS, or even a distinct condition. The Working Group accepts that some patients' presentation and symptoms align more closely to the original clinical description of ME." (3.4.1)

To complete its task, the Working Group side-stepped the issue:
"We recognise that no current terminology is satisfactory, so in line with our original terms of reference [MERGE emphasis] we have used the composite CFS/ME for the purposes of this report, acknowledging that CFS is widely used among clinicians and ME among patients and the community." (3.2)
"For how much longer will anomalies in nomenclature complicate and obscure clinical care?"

The issues surrounding the establishment of CFS as a diagnostic category, and the inaccurate and biased characterisations of CFS that have subsequently arisen, have been well reviewed by Jason et al (1997):

"Over the past ten years, a series of key decisions were made concerning the criteria for CFS diagnosis and the selection of psychiatric instruments, which scored CFS symptoms as medical or psychiatric problems. At least some of these decisions may have been formulated within a societal and political context in which CFS was assumed to be a psychologically determined problem (Manu et al. 1988). Many physicians and researchers believed that CFS was similar to neurasthenia and that CFS would eventually have a similar fate once people recognised that most patients with this disease were really suffering from a psychiatric illness. Psychiatrists and physicians have also regarded fatigue as one of the least important of presenting symptoms (Lewis & Wessely, 1992). These biases have been filtered to the media, which has portrayed CFS in simplistic and stereotypic ways... One major consequence is that many CFS patients feel dissatisfied with their medical care... and have gone outside traditional medicine to be treated for their illness...

"A significant complicating factor in understanding the dynamics of this illness is that there are probably different types of illnesses now contained within the CFS construct... We believe that it is crucial for CFS research to move beyond fuzzy recapitulations of the neurasthenia concept and clearly delineate precise criteria for diagnosing pure CFS and CFS that is comorbid with psychiatric disorders. It is also necessary to better differentiate CFS from other disorders which share some CFS symptoms but are not true CFS cases."

One of the most poignant sentences in the report is:

"The severely ill reflected strong loathing of the name CFS because fatigue is often not perceived to be their main problem; ME is a preferred term by many." (2.3.1.1)

For these and other patients, the question is how much longer anomalies in nomenclature will be allowed to complicate and obscure clinical care. Given that the term `CFS' most probably groups different kinds of patients under one umbrella, management recommendations are likely to be inadequate and probably misleading.

3.4 Choice and interpretation of best management strategies

It is important to realise that the Working Group was empowered to identify evidence for "management strategies" not treatments, since it is clear that none of the forty-four randomised clinical trials found and reviewed (Whiting et al, 2001) supplies convincing evidence of treatment efficacy for a specific symptom or condition. At first sight, the process of identification of `useful' management strategies appears clear:

"Where research evidence exists we have been guided by it. (1.0) ...We used a trident approach to review and synthesise three lines of evidence: research findings, patient reports, and clinical opinion... Members of the Working Group expressed widely differing opinions on the potential benefits and disadvantages of these approaches. However, we agreed that all could be considered as management options in line with general principles outlined here... The Working Group agreed that there is no cure for CFS/ME but identified three specific strategies as potentially beneficial in modifying the illness: graded exercise, cognitive behavioural therapy, and pacing." (4.4.2)

Though the report contains several caveats about all three `management strategies' - perhaps as a sop to the non-biopsychosocial opinion on the committee (i.e., patients and carers) - it is the choice of these which, in the end, provides justification for the existence of the Working Group and the money spent (including that provided by the Linbury Trust) on the CMO report. Yet how each of the three strategies was determined to be "potentially beneficial" is not as clear as it might appear. As regards the published evidence, the thorough review by Whiting et al (2001) state that it is very difficult to draw overall conclusions (from the forty-four randomised clinical trials) since very little information is available on baseline functioning. Most of the interventions were evaluated in only one or two studies, so the validity of generalising the findings is limited. Since there are few patient reports favouring cognitive behavioural therapy, and a sizeable proportion of patients feel that graded exercise therapy worsens their condition, the inference must be that the major recommendation for the use of cognitive behavioural therapy and graded exercise therapy was clinical opinion, the only other source of evidence left to the Working Group. If this is the case, then the professional composition of the Key Group was the crucial factor in determining the strength of recommendation for particular "potentially beneficial" management strategies.

It is also important to realise that research funding is critical to whether or not evidence is available. There are indications that psychiatric and psychological research groups conducting trials of cognitive behavioural therapy and graded exercise therapy have been particularly well-funded (Abbot & Spence, 2002); hence, the forty-four trials available for analysis by Whiting et al (2001). This funding bias is itself worthy of examination as it informs us that the research agenda in CFS/ME has been driven, in the main, by a relatively small number of clinicians with a professional interest in exploring biopsychosocial models of illness. These clinicians were proportionately well-represented within the Working Group.

3.4.1 Cognitive behavioural therapy

The issues surrounding the true usefulness of cognitive behavioural therapy for CFS/ME patients have been widely discussed (e.g. Lancet 2001; 358: 239-41) but can be summarised as follows:

Of the forty-four randomised clinical trials identified, only five involved some variant of cognitive behavioural therapy, and of these, three had a `positive' result and two a `negative' result. Two of these trials used the Oxford criteria which greatly limits the applicability of the findings as far as ME and CDC-defined `CFS' is concerned. Dropout rates were high - 40% in the active arm (vs. 20% in the control) of the flagship trial on cognitive behavioural therapy by Prins et al (2000). As Whiting et al (2001) state in their review: "Dropout rates may be an indication of the acceptability of an intervention" and "cognitive behavioural therapy may be acceptable to only a small number of patients, limiting generalisability."
As is the case with most clinical trials, the results cannot be extrapolated to apply to the most severely ill (up to 25% in CFS/ME), nor to children or young people. Both categories having been excluded from these trials.

While cognitive behavioural therapy most likely has some role in helping some patients to better cope with their symptoms until a cure is found, this role is limited (as it would be with cancer patients) and non-curative.

Cognitive behavioural therapy is expensive and, with such a variable outcome, the cost-benefit ratio is problematic.
As well as the limitations of the clinical trials in CFS/ME patients, there are doubts even among professionals about the specific efficacy of cognitive behavioural therapy. As a recent review commented:

"...the foundations on which it rests are not as secure as some of its proponents would have us believe." (Holmes, 2002).

"The foundations of cognitive behavioural therapy are not as secure as its proponents suggest."
Though the CMO report states that "application of a cognitive behavioural model to CFS/ME has been found successful in most patients in the trials" (4.4.2.2), this bald statement is almost certainly untrue: of five randomised controlled trials, two were negative, dropouts were high, and some `improvements' were seen in the control groups, indicating that not all improvement can be ascribed to CBT. The same section of the report contains a remarkable statement:

"The Working Group accepts that appropriately administered cognitive behavioural therapy can improve functioning in most patients with CFS/ME who attend adult outpatient clinics." (4.4.2.2)

This is a masterful piece of drafting which skilfully suggests great benefits of cognitive behavioural therapy while leaving several exits in case of attack. What is "appropriately administered" cognitive behavioural therapy? What aspect of `functioning' is meant? How can the Working Group accept that `most' patients improve on the basis of the extrapolation of the results of three positive and two negative trials to the whole population of CFS/ME patients in the UK?

There are several quotations in the report which - probably unwittingly - go to the heart of the matter:

"Cognitive behavioural therapy for people with CFS/ME is currently unavailable or very difficult to obtain in much of the UK." (4.4.2.2)

"There was disagreement among clinicians as to the precise value and place of cognitive behavioural therapy, which partially reflected the varying models of the therapy and disease." (4.4.2.2)

"We also noted that misunderstanding, misplaced concern, and poor practice in this area could potentially undermine the beneficial application of this therapy or its principles in patients with CFS/ME." (4.4.2.2)

"In one patient-group survey, only 7% of respondents found the therapy [CBT] `helpful', compared with 26% who believed it made them `worse'. The remaining 67% reported `no change'." (4.4.2.2)

As these quotes help to illustrate, cognitive behavioural therapy is non-curative (Wessley, 2001); is expensive and time-consuming, and beyond the resources of Health Authorities to fund; has an irrecoverably poor reputation among ME patients, especially the severely ill whom it incenses; has been found helpful by only a small minority of patients surveyed; and requires skilled therapists who need the consent of malleable patients rather than irate unwilling ones. As a recent commentary in the British Medical Journal stated:

"Until the limitations of the evidence base for cognitive behavioural therapy are recognised, there is a risk that psychological treatments in the NHS will be guided by research that is not relevant to actual clinical practice and is less robust than is claimed." (Bolsover, 2002).

Or, as one patient has said, cognitive behavioural therapy is "not curative, not cheap, not accepted, and not the answer for everyone."

3.4.2 Graded exercise therapy

Graded exercise therapy was the other "potentially useful" therapy identified by the Working Group on the basis of the three positive clinical trials out of the forty-four identified. The limitations of these trials have been discussed in depth elsewhere (BMJ 1997; 315: 947 and electronic responses to BMJ 2001; 322: 387), but the main points can be summarised as follows:

The success of randomised controlled trials depends on strict comparability of control to treatment groups. In these trials there was not the same contact with the controls and patients, raising the possibility that factors other than treatment were involved in the "positive" outcome.

All three trials consisted of patients classified by the Oxford criterion which does not diagnose ME or the CDC-CFS criteria (Fukuda et al, 1994) exclusively. The weakly-positive trial results may reflect this bias, have little relevance to CFS/ME patients, and have no relevance to the large numbers of severely affected or young sufferers. Graded exercise therapy involves a patient-motivation component to encourage compliance with the exercise regimen. However, the true usefulness of such programs is by no means clear (Harland, 1999).

Its use is predicated on the belief that deconditioning is a factor in the perpetuation of illness in CFS/ME patients. However, there is good evidence that deconditioning is not a significant factor (Brazelmans, 2001; Van der Werf, 2000) and that it cannot account for delayed post-exertional symptoms or the documented changes in muscle metabolism (Lane et al, 1998; Lane, 2000).

None of these is successfully dealt with in the CMO report, though some limitations are alluded to:

"One key controversy that exists over graded exercise rests on whether the nature of the treatment is appropriate for the nature of the disease, at least in some individuals. Existing concerns from voluntary organisations and some clinicians include the belief that some patients may have a primary process that is not responsive to or could progress with graded exercise, and that some individuals are already functioning at or very near maximum levels of activity." (4.4.2)

"Voluntary organisations, as well as the Sounding Board events, note that graded exercise therapy can be effective in some individuals, but substantial concerns exist regarding the potential for harm." (4.4.2.1)

Fortunately, some hard evidence from patient surveys is shown in the Working Group's report, albeit in Annexe 3. This showed that of 1,214 patients using graded exercise therapy, 34% found it helpful but 50% (610 patients) reported that it made them worse. Graded exercise therapy had the greatest number of `worse' reports of any therapy.

Clearly, as a management strategy, graded exercise therapy has its limitations for CFS/ME patients:
"Best practice in this area indicates that the initial stages of any graded exercise programme should only be carried out by therapists (i.e., occupational therapists, physiotherapists, exercise physiologists, sports therapists, etc.) who have the necessary expertise to manage CFS/ME patients."
(4.4.2.1) At present, very few therapists are available with such expertise.

3.4.3 Pacing

In contrast with the two professionally-dictated interventions, pacing has been included as a `management strategy' in response to patient experience - an example (some might say) of patients voting with their feet. Pacing allows patients to choose their own acceptable level of activity in accord with their fluctuating symptoms. It accepts that in the rehabilitation of sufferers, rest and relaxation also have an important role to play (Shepherd, 2001). The report clearly states the rationale for pacing:

"Clinical wisdom suggests that management of limited energy and supervision of any increases in physical or mental activity are an essential part of ongoing care for individuals with CFS/ME." (4.4.2)

"A survey of more than 2,000 members of a voluntary organisation (Annexe 3, section 3) who were or had been severely unwell showed that 89% of group members found pacing `helpful'." (4.4.2.3)

While pacing is intuitively sensible, its status as a clinical management strategy chosen after three years of deliberation by a Working Group is debatable, and there is a lingering suspicion that it has been recommended by the Working Group only as a concession to patient-based opinion. Whether sufferers will be allowed by healthcare professionals to choose this "recommended" therapeutic strategy in preference to psychological strategies is an open question. Indeed, almost as soon as the Working Group's report was published, an item in the British Medical Journal commented: "The clinicians argued that the psychosocial side of the condition should have had greater emphasis and were concerned that `pacing'... was included as a form of treatment," and quoted one professional as saying that "...doctors would not accept pacing just because it was recommended in the report" (Eaton, 2002).

3.4.4 Conclusions about the choice of management strategies

The preamble to the CMO report was explicit in its aims:

"Throughout, we have aimed where possible to base our commentary and recommendations on the best quality evidence, and from a range that includes randomised controlled trials and clinical anecdote. In the absence of research evidence to inform many issues, the bulk of the report is derived from a synthesis of patients' and clinical experience. Where some data exist, albeit incomplete and not fully agreed, we considered the trident approach together with the likely resource implications to inform our conclusions." (1.3.3)

How far have these aims been achieved? By conventional standards of literature reviewing, formal evidence for the use of cognitive behavioural therapy, graded exercise therapy and pacing is rudimentary. The fact that a few more clinical trials exist for cognitive behavioural therapy and graded exercise therapy than for any other intervention merely reflects the funding support which the interventions attract in the UK (Abbot & Spence, 2002).
Patient evidence suggests that a small subgroup of patients might find either cognitive behavioural therapy or graded exercise therapy helpful (7% and 34% respectively) - possibly reflecting the heterogeneity of the patient grouping inside the construct `CFS' - but that a substantial proportion (93% or 66% of patient responders, respectively) either find them ineffective or harmful. Pacing is nothing more than a commonsense approach enforced on most patients by their circumstances, and can hardly be described as a therapeutic management strategy. To use an analogy, pacing could describe the ability of an amputee to hobble around in difficult circumstances:
a "therapeutic management strategy", however, might include a new prosthesis individually designed. Strangely, in a recondite section (but not in the easily-accessible overall conclusions) the report itself admits the truth:

"Review of the evidence highlights the lack of good quality research to support effectiveness of various therapies. Patient responses suggest that no approach is universally beneficial and that all can cause harm if applied incorrectly." (4.0)


Continued tomorrow Feb 4

MERGE is the Myalgic Encephalomyelitis Research Group for Education and Support