When I first trained in science, medical researchers used valid definitions, analysed their data with appropriate tests, and applied a dispassionate objectivity to their results. The best journals operated a peer review system, which upheld high standards by checking the text for errors and challenging unsubstantiated claims. They didn't always get it right, but mistakes were usually corrected and there was no evidence of systematic bias.
The most prestigious of the British publications were the British Medical Journal (BMJ) and the Lancet. As readers, we knew that at least two experts had gone through every article with a fine toothcomb, and we therefore assumed that the information was generally accurate and reliable.
When I read articles relating to my previous work as a researcher (endocrinology/gynaecology), this is still the case. Indeed, I cannot remember the last time when I saw anything in either journal which was at odds with what I already knew or had read elsewhere. However, that's not true with respect to articles on chronic fatigue syndrome (CFS).
In this essay, I will describe the misrepresentation of the illness and the research on CFS, the selective discussion of the evidence, the obvious preference for psychological explanations as well as the inaccurate portrayal of the patients, their beliefs and their behaviour. What is happening is unacceptable. There are rules which limit what scientists can say and write, and these journals are not adhering to them. With respect to other illnesses, editors do not allow doctors to ignore important and scientifically sound research which just happens to undermine their arguments. They do not allow writers to misrepresent and misinterpret their colleagues work or to put a 'spin' on the evidence so as to make it conform to their own preconceptions. Similar rules apply in other professions. As The Hon. Mr. Justice Gray noted at the end of the court case involving the Holocaust denier David Irving: "an objective historian is obliged to be even-handed in his approach to historical evidence: he cannot pick and choose without adequate reason" (para 13.24). As I will show, some doctors not only 'pick and choose' when discussing CFS but in doing so, they invariably trivialize the illness and exaggerate the role of psychiatric factors or the effectiveness of psychological interventions.
The situation in Britain is further complicated by another practice. Most editors of medical journals respect the need for balance and are happy to publish a wide range of opinions and theories. However, when it comes to the journals like the BMJ and Lancet, you won't find much information which challenges the popular psychological explanations of CFS. In essence, they have excluded scientists like myself from presenting our data and taking part in the debate. They still publish our letters and we continue to have access to specialist publications, but you will not find any papers reporting our research, nor any news items which undermine their preferred approach.
As a result of these editorial policies, most doctors in Britain probably have a very limited and distorted view of the illness. Their narrow frame of reference must undermine their understanding of the condition and their ability to make truly informed decisions about treatment. In the short-term, the solution is to warn them of the bias and ask them not to rely on the Lancet and BMJ for their information on CFS. In the long term, we should encourage more journals to set up independent committees and submit complaints when appropriate. If that doesn't solve the problems, then we may need the help of a dedicated Ombudsman with legal backup.
In the meantime, here is just some of the evidence which supports my thesis. Due to a lack of space, I will focus on the one-sided discussions, and the tendency of editors to overlook misrepresentation and publish information which is unreliable or inaccurate.
Although
most cases of psychologisation involve British psychiatrists, American
physicians can be just as biased! One of the most influential is internist Peter
Manu, who regards CFS as a
functional somatic syndrome (1998). In his recent editorial, (Manu 2000), he
began by down-playing the research showing upregulation of the
anti-viral RNase L pathway (De Meirleir et al 2000) and by ignoring all
the other evidence which supports the ' immune dysfunction model' (e.g. Cannon
et al 1997, Landay et al 1991, Ojo-Amaize et al 1994). Then he discussed
the 'key findings' supporting a psychological explanation, but he chose to
ignore the methodological flaws (Jason et al 1997) and misrepresented some of
the relevant research. For instance, he cited a study by Sharpe et al
(1992) which had apparently shown a relationship between "a viral belief at
the time of diagnosis" and functional impairment "1 year later". However, the belief in a viral cause was actually assessed at
follow-up, so one can not deduce that the former was causally related to
the latter.
Manu also referred to three other papers as showing a correlation between viral beliefs and poor outcome. They included the study by Chalder et al (1996) which revealed that a belief in a psychological, not physical, cause was predictive of fatigue scores at follow-up.
The second study cited by Manu actually reported a correlation of 0.1 between physical attributions at initial assessment and fatigue at follow-up (Vercoulen et al 1996). This was clearly not significant. There were also other findings, but nothing provided strong support for Manu's argument.
The only reference which supports a relationship between a belief in a viral cause and poor outcome is the Australian study by Wilson et al (1994). Interestingly, these authors admitted that they had misdiagnosed one in five of their subjects and suggested that the sample as a whole may not have been "representative of the whole spectrum of patients with the CFS".
Suffice to say, Manu did not allude to the studies which found no relationship between beliefs in a physical cause and poor outcome (e.g. Deale et al 1998, Lawrie et al 1997). Instead, he described further support for a psychological explanation, namely "the success of randomized trials of cognitive behaviour therapy and graded exercise". He cited two studies for each (e.g. Deale et al 1997, Fulcher and White 1997). However, he omitted to mention the two controlled trials which failed to support the effectiveness of CBT for CFS ( Friedberg and Krupp 1994, Lloyd et al 1993) and the reports of relapses following overexertion (e.g. Lapp 1997).
To be fair, this editorial was balanced by another one in the same issue. However, the point is that Manu misrepresented the research on the psychological aspects of CFS and that this was published in a respected medical journal. To counter it with a discussion of the evidence of disease isn't enough. The accounts of the psychological research should be evaluated in the same way as the articles on immunology and virology. The same rules should apply. Psychological medicine shouldn't be a second class speciality where you can write anything you like. If we allow psychiatrists and psychologists to manipulate evidence and rely on generalisations, we can hardly criticise complementary practitioners who do exactly the same.
Scientists have more freedom in books than in medical journals but since the peer review system doesn't appear to operate in relation to CFS, it makes little difference. Here is another example where the writer seems far from impartial. On page 22 of his book, Manu described two studies on magnetic resonance imaging (MRI). He rightly concluded that most abnormalities have been non-specific and that MRI therefore "does not have a substantial clinical utility as a screening test in the differential diagnosis of CFS".
At
the time of writing, this was a perfectly fair assessment of the existing
literature on MRI. However, he did not go on to mention the much more
significant (and consistent) data which have been obtained using SPECT.
Did Manu not know about that research? Yes, he must have done. One of the
two studies he cited in relation to MRI was actually a comparison of the two
techniques (Schwartz et al 1994). In fact, that study revealed abnormalities in
81% of the SPECT scans, and found that "the number of... abnormalities
appeared to correlate with clinical status".
Manu also ignored another paper on SPECT, written by the same authors and
published in the very same issue. It provided further support for the theory
that some types of CFS are the result of ongoing disease (cf. Costa et al 1995,
Tirelli et al 1998).
Here are two examples of misleading information published in the BMJ. They come from Tony Delamothe, then (and possibly still) assistant editor (1994). In a typically unhelpful and biased article, he implied that patients with CFS had manipulated the World Health Organisation (WHO) and got them to "include myalgic encephalomyelitis under diseases of the nervous system" in ICD-10. The truth is that patients were not consulted by the WHO and played no part in their decision to include the term in ICD-10. I know this, not least, because I rang the WHO in Geneva and asked them.
Delamothe also alleged that "supporters of myalgic encephalomyelitis as a discrete entity with an organic cause... landed a Myalgic Encephalomyelitis Act on to the British statute books, requiring an annual report to be made to parliament on its causes, effects and treatment." This again is untrue. There's no ME Act. A Ten Minute Rule Bill was introduced by Jimmy Hood MP in the 1987-1988 session but failed. He then tried to introduce a Presentation Bill in the 1990-1991 session but this also failed.
Apart
from showing the willingness to publish totally unreliable evidence,
these examples also illustrate the final point I wish to make. The errors
which one finds in the British journals with regard to CFS almost always support
the psychiatric explanation or the negative stereotype of the patients.
They do not occur at random (see also Henningsen and Priebe 1999, Wessely
et al 1998). If they were genuine
mistakes, one would not expect to find such consistency.
The situation I have described cannot be attributed to differing interpretations of the research, nor am I suggesting that certain people should not be allowed to express their opinion. To paraphrase the Hon. Mr. Justice Gray, an objective scientist is obliged to be even‑handed in his approach to scientific evidence: he cannot pick and choose without adequate reason. Does the psychologisation of CFS constitute an "adequate" reason? Of course it doesn't. In my view, doctors should not omit, disregard or dismiss credible evidence, and skew evidence to bring it into line with their personal beliefs. They can do that when writing as private individuals in non‑scientific publications, but not in their capacity as medical professionals. Similarly, papers should be assessed by independent reviewers, and judged according to the quality of the evidence. To reject all articles which challenge a psychological explanation is incompatible with the scientific process and could be construed as an abuse of trust. And there should be no place for writers who reinforce inaccurate stereotypes and encourage discrimination and prejudice. There is nothing which justifies these practices. Bias and misrepresentation in medical publications disempowers, confuses and misleads doctors. As a result, they may find it hard to understand what their patients are experiencing, and not be able to provide the help they need and deserve.
Returning to CFS, I cannot think of an adequate reason why writers should continue to ignore the differences between the subgroups. One might as well claim that migraines are not worth differentiating from ordinary tension headaches. Nor can I think of a good reason why colleagues are still advocating graded activity as a universal treatment, given the evidence that it is neither appropriate nor effective for most people with uncomplicated, strictly-defined CFS (Jason et al 1997). I don't understand the obsession with fatigue and the inactivity theory, when the reality is so different. I just don't see how this blinkered approach helps patients. How can they trust orthodox medicine when some of its practitioners act more like spin doctors than scientists?
This issue is not about a difference of interpretation or about wanting to avoid the stigma of mental illness. It's about scientific intolerance and a lack of respect for colleagues, it's about manipulation and about activities which are inconsistent with the Hippocratic Oath.
I have no doubt that it will take years for this problem to be taken seriously. Usually, the first response to any criticism of one's colleagues is disbelief. This is invariably followed by denial and eventually, by wholesale rejection. In medicine and psychology, we still respond to this kind of issue by shooting the messenger and by closing ranks. After all, we know how pressure can get to you and how easy it is to make mistakes! Only when a much respected, usually male member of the establishment makes a similar claim, will most of us even consider the possibility that some of our colleagues are being, shall we say, unprofessional? Aside from that, we simply don't want to believe that journals like the BMJ can no longer be trusted to provide reasonably authoritative and reliable information. So I personally expect to meet resistance and patients should too. In the meantime, there is no reason why the issue should not be discussed.
To
sum up, I believe that as people we have freedom of speech, but as scientists,
we do not have the freedom to mislead!
Copyright AAA ©
Not to be reproduced without permission.
All rights reserved. This article may not be reproduced
See the full copyright notice.
Be sure to see the many other valuable articles at our Main M.E. Home Page
