Sunday, April 24, 2011

Some theoretical factors to consider when analysing research findings in CFS

Some theoretical factors to consider when analysing research findings in CFS

Some researchers appear to think it worthwhile to mention whether CFS research participants are in receipt of state benefits or insurance payments, are awaiting an appeal to settle a claim, are members of a support group and what their 'beliefs' are about the cause of their illness.  I imagine such researchers consider these 'secondary gains' and/or motives for remaining, or appearing to remain ill that could influence the outcome of research.  The following are some other factors that might be worth considering when weighing-up research results.

1. Coping effect of CBT unrelated to treatment of illness

A selling point used by some, is to inform people with CFS that CBT is used for other illnesses; i.e., MS, diabetes, etc.  CBT helps some people with those illnesses to cope better but it does not cure the disease.  Therefore a certain degree of coping effect must be deducted from all reports or measures of improvement with CBT for CFS; because at least some proportion of any improvement recorded only relates only to coping and not to treatment for the illness.  I imagine that some Wessely school theorists would claim that CFS is only in fact; an illness of not coping and therefore improving coping is affective or even curative.  However, the evidence from biomedical research and the results of the PACE Trial, suggest this is an erroneous belief.

2. Inclusion of non-target patients in research.

It is notable that in their retrospective analysis of their 2008/9 patients by applying the Fukuda criteria for Chronic Fatigue Synfrome (CFS), Newton et al (2010) found that 40% of all fatigued patients referred to the Newcastle CFS clinic were subsequently found not to have CFS.  47% of these were found to have a chronic disease and 20% a primary sleep disorder and 15% psychological/psychiatric illnesses causing their illness.  The entry criteria in the PACE Trial was CFS diagnosed by the Oxford criteria which does not diagnose CFS as many authorities understand it.  It might be a sensible precaution to view outcome data sceptically to allow for the inadvertant inclusion of non-CFS, psychologically ill participants and those deconditioned by a sleep disorder.

3.  Use of subjective measures.

In the PACE Trial CBT and GET explain CFS symptoms with theories of fear avoidance, abnormal illness beliefs, deconditioning, etc.  Yet the PACE Trial used almost entirely subjective measures such as quality of life, symptom and fatigue questionnaires to assess baseline and treatment effect.  So although the theories underpinning the interventions on trial imply that people with CFS are incompetent to assess and interpret their symptoms; the PACE Trial used these same people's assessment as evidence in a controlled clinical trial.  It is notable that patients wanted more objective measures in the PACE trial - but the researchers actually reduced the use of objective measures.

4. Client deference and subjective measures

When research uses a one-to-one, client/therapist scenario a therapeutic alliance occurs in which the balance of power is unequal.  How participants rate subjective questionnaires may be influenced by how a client feels about their therapist, whether they like, dislike or possibly even fear them.  In "Clients' Deference in Psychotherapy", Rennie (1994) identified 348 instances of client deference to the therapist occurring in just 16 therapy sessions.  These most commonly related to the client's, "Concern about the therapist's approach", followed closely by "Fear of criticising the therapist".  Rennie also noted that therapy provided free of charge (as in research) was a factor likely to engender deference.  In the context of CFS psychosocial research, this suggests that participants (who may have been subjected to pressure to doubt their own experience and rationality) may rate subjective measures deferentially; i.e., in a way that they believe the therapist would approve of.

5.  Natural improvement not associated with treatment.

The National Institute for Clinical Excellence in association with Royal Society of Medicine (2002) state: "It is generally recognised that prognosis is variable.  Many patients improve quite quickly. However, in those who do not improve quickly, the illness can persist for a long time.  The prognosis tends to be worse for severely ill patients than for less severely ill patients."  This shows that research into CFS that does not use a representative sample (i.e. including 25% severely affected and long-term patients) cannot be extrapolated to the patient population as a whole.  It also suggests that participants who have been ill for (guestimate) less than 2 years are potentially confounding - because many of these are expected to naturally improve.  In the PACE Trial, 45% of participants in the control group spontaneously improved.

6. Chronic Infectious Mononucleosis.  (This factor might combine with or be separate from 6 above)  Professor White, Principal Investigator of the PACE Trial was a co-author of Candy et al (2002) which states, "The prevalence of prolonged illness after IM varies, from 1.5% to 56%.  The symptom consistently reported to have longest duration was fatigue.  One of the two largest prospective studies found that, six months after onset of IM, physical fatigue was present in 40% of subjects and 9% to 22% had a CFS" (the latter was also White's research).  Candy et al also remark, "Physical deconditioning may play a key role in causing prolonged symptoms".  According to Thompson et al (1969, cited by Candy et al), 56% IM cases reported fatigue at 6 months, and 16% at 11 months.  Therefore, many chronic IM patients spontaneously improve and might be expected to respond to rehabilitation.  The inclusion of chronic IM patients could confound research of interventions in CFS by biasing results towards improvement.  Research that does not state that such patients have been identified for sub-grouping or exclusion might require more sceptical examination.

7. Willingness to participate

Some patients may feel reluctant to refuse participation in research if their medical professional suggests they consider taking part.  Given the difficulty that people with CFS sometimes have in getting medical support it is not difficult to imagine that some may be very reluctant to appear to be going against their medical professional's wishes when research participation is mentioned.  Then, once enrolled, in the participant's mind there might be a sense of a parallel relationship influencing them; as if their medical professional were looking over their shoulder, as it were.  Aside from the possibility that this might constitute a form of invisible coercion; it could also influence how a participant reports the effect of interventions.  They may not want to appear uncooperative or critical, but if they don't improve they might compensate for the conflicting pressures they are feeling by indicating modest improvement.  This could be face-saving for both the referring doctor and the patient.  The latter, because they do not have to confront their vulnerability or how this could have led to their reluctant participation.  If correct, this would indicate that some participants might rate subjective measures in a political context.  Ironically, the Wessely school researcher's attention to benefits claimants and support group members seems to be suggesting precisely this type of problem; but in their case they see the problem as resulting in 'worse' results for their therapies; whereas some participants that feel coerced might be distorting results by making them look better than they really are.

Some researchers give consideration to 'Double-Vulnerability'.  This occurs when potential participants have more than one disempowering factor making them especially vulnerable; so that what might be considered acceptable encouragement and  persuasion for others is felt as, and operates like coercion.  Due to the considerable disability and losses often seen in CFS plus the disenfranchising nature of the diagnosis with stigma and prejudice, people with CFS should be considered to have Double-Vulnerability.  Methods of recruiting participants with the illness require careful consideration.

8. Participant conflicts of interest.  When Wessely school researchers report participant factors such as benefits or insurance claims and support group membership (which they associate with negative outcome) they do not appear to question: why did patients with apparently conflicting interests enrol in the first place?  Did they join just to spoil the research or is there some form of coercion going on?  Or perhaps: do patients who are informed and supported by fellow patients feel empowered to be more critical in their analysis of a treatment?  And maybe: what does the additional stress of a benefits claim do to a vulnerable and disabled patient's illness when it threatens support that they are dependent upon?  These rational questions do not appear to receive consideration.

9. Reporting research results. The PACE Trial results have shown quite convincingly that Graded Exercise Therapy (GET) and Cognitive Behavioural Therapy (CBT) are not effective interventions for patients with Chronic Fatigue Syndrome (CFS).  Only 14% -16% more participants in the GET or CBT groups improved compared to the control group.  The results of this major controlled clinical trial do not support the recommendation of either of these therapies.  Yet some in the media have reported this £5 million experiment as though it indicated that GET and CBT are useful; apparently because of the way the results were represented to them.  It seems that the press conference paid little attention to the failure of the therapies to make any significant difference over and above the Control Group, but focussed on the difference between GET or CBT when compared to Adaptive Pacing Therapy (APT).  This resulted in some of the media reporting the research as though it suggests that people with CFS simply need to do more exercise; when what the results actually show is that such behaviour modification makes no difference to the majority of patients.

10. Competing interests don't just mean 'money'.  There are sound reasons for the requirement for researchers to declare 'conflicting interests'.  These sometimes relate to financial matters; yet vested interests are not limited only to 'money'.  Professional credibility could be equally important.

Some of the researchers in the PACE trial espouse theories derived from what is referred to as the - 'Wessely school', in that they are proponents of Professor Simon Wessely's theories about CFS.  Professor Wessely's theories of CFS include hysteria and phobia.  In 1994 he stated; "I will argue that ME is simply a belief, the belief that one has an illness called ME".

Wessely school theories underlie the professional practice of some involved in the PACE Trial research, as well as 2 of the therapies being tested.  If the research results contradict Wessely school theories it might be humiliating to those whose professional practice is based on these theories.  It could show that they have wasted time and money and involved participants in futile research.  It might indicate that theories about, and treatment of patients has been inadequate or even completely inappropriate.  To some, this might appear to be a competing interest.

Peter Kemp


Candy, B., Chalder, T., Cleare, A., Wessely, S., White, P.D., Hotopf, M. 2002. Recovery from infectious mononucleosis: a case for more than symptomatic therapy?  A systematic review.  British Journal of General Practice.  October.

National Institute for Clinical Excellence in association with Royal Society of Medicine Press. (2002) Interventions for the Management of CFS/M.E.  Effective Health Care.  University of York: Vol. 7,  Number 4.

Newton JL, Mabillard H, Scott A, Hoad A, Spickett G. (2010).  The Newcastle NHS Chronic Fatigue Syndrome Service: not all fatigue is the same.  Journal of the Royal College of Physicians Edinb. 2010 Dec;40(4):304-7.

Rennie, David, L. 1994. Client's Deference in Psychotherapy.  Journal of Counseling Psychology. 41.4.427-437.

Thompson, DS., Godleski, J., Herman, S. 1969. Prognosis post infectious mononucleosis.  Journal of American College Health. 17:453-457.

Simon Wessely.  1994.  Microbes, Mental Illness, The Media and ME: The Construction of Disease.  9th Elliot Slater Memorial Lecture, Institute of Psychiatry, London, 12 May 1994.

No comments: