New research


Summaries and comments by Dr. Ellen Goudsmit

Archivist and Chartered Health Psychologist, London.


Bazelmans E, Bleijenberg, G., van der Meer, JWM and Folgering, H. Is physical deconditioning a perpetuating factor in chronic fatigue syndrome? A controlled study on maximal exercise performance and relations with fatigue, impairment and physical activity. Psychological Medicine, 2001, 31, 107-114. 

Twenty patients with CFS (CDC criteria '94) were compared with 20 matched neighbourhood controls on a maximal exercise test with incremental load. The mean duration of illness was 3.2 years and all had severe fatigue (CIS score).  Measures included heart rate, blood pressure, oxygen consumption and saturation, carbon dioxide consumption and blood-gas values (of arterialized capillary blood).

Physical fitness was quantified as the difference between actual and predicted ratios of maximal workload versus increase of heart rate. Fatigue, impairment and physical activity were assessed to study its relationship with fitness.

There were no statistically significant differences in physical fitness between the CFS patients and the controls. Indeed, nine of the patients were fitter than their respective control.  There was a negative relationship between physical fitness and fatigue in both groups.

In the CFS patients, there was a negative correlation between fitness and impairment (p=.027)and a positive correlation between fitness and physical activity (p=.039). Some CFS patients failed to achieve a "physiological limitation at maximal exercise" reaching, for example, 70% of their predicted workload compared to 83% in the controls. However, there were no differences in the heart rate scores during exercise.

The researchers conclude that "physical deconditioning does not seem a perpetuating factor in CFS".

[Ed. Note: This is high quality research with noteworthy findings (especially given it comes from adherents of the CBT model). In the first study to look at the relationship between fitness and impairment, the results challenge the most important assumption of the model, i.e. that the fatigue is largely maintained by reduced physical fitness. The findings are consistent with those of van der Werf et al, (J. Psychosom Res 2000, 49, 373-379) and the epidemiology of CFS and ME. De Becker et al (Arch Intern Med 2000, 160, 3270-3277) found reduced exercise capacity in a more strictly defined sample, but this provides no support for the CBT model as the results were related to disease severity and similar to those documented in other chronic conditions.

Bazelmans et al’s paper is not perfect and there are some minor flaws, e.g. the small sample, the lack of information on depression and how this might have affected the findings. Nevertheless, this study and that of van der Werf et al are a cut above the rest.  We know how fatigued the patients were, we have reliable data concerning their activity levels (some researchers just ask subjects for an estimate), and many readers will appreciate the balanced, evidence-based discussion, devoid of superfluous speculation and conjecture. I don’t quite understand why 9 patients with severe fatigue were fitter than the healthy controls and this study will need to be replicated using a larger sample. However, the reports were a joy to read and in my view, both are major contribution to the literature!]

Pope, J and Woodcock, A. Chronic fatigue syndrome: causal attributions, coping, psychological adjustment and anti-depressant adherence. Proceedings of the British Psychological Society, 2001, 9, 1, 73.

This is a report of a study on 61 members of a self-help group for people with ME (MESH). Measures included the IMQ to assess coping strategies, the HAD to assess anxiety and depression, and the IPQ to assess perceived causes.

It was found that 73.3% of the patients attributed their illness to a germ of virus, 68% to an 'immune system problem' and 67.2% to stress. Over 70% felt that depression was not a major cause of ME.

The findings did not support the researchers’ hypothesis that physical attributions were negatively related to maintaining activity and positively to focusing on symptoms and accommodating to the illness (IMQ). There was also little support for the hypothesis that emotional attributions would relate positively, and physical attributions negatively to anxiety and depression.  (Only those who believed depression to be a major cause of their illness were more anxious). Focusing on symptoms was weakly but significantly related to higher depression levels (r=.36 p<0.05) but not to anxiety, thus partly supporting the third hypothesis. Compliance as regards antidepressants was not more common in those making psychological attributions.

Butler, JA., Chalder, T and Wessely, S. Causal attributions for somatic sensations in patients with chronic fatigue syndrome and their partners. Psychological Medicine, 2001, 31, 97-105.

This study involved 50 patients with CFS, 46 of their relatives and 50 controls from a fracture clinic. Measures included the HAD, three (?) MOS short form subscales, and the modified Symptom Interpretation Questionnaire. They were asked to decide whether a number of common symptoms were likely to have a physical (somatic) or psychological cause, or whether they were the result of external events. There were two conditions: the first asked for an assessment as if the person had the symptom; the other asked for a response assuming that the partner had the symptom.

CFS patients were more likely to make somatic attributions for the listed symptoms (including prolonged headache, sweating, sudden dizziness, stomach upset, hand trembling, trouble sleeping, lost appetite) than people attending the fracture clinic. The same tendency was found for their relatives when asked to imagine that the listed symptoms occurred in the person with CFS.

Sixty-one per cent of the patients ascribed their own illness to a physical cause. More than 90% of the fracture clinic group blamed an accident or fight for their symptoms.

The researchers suggest that a tendency to make somatic attributions for symptoms "may be a vulnerability factor for the development of CFS". They discuss the potential value of CBT.

[Ed. note: This is one of the strangest reports I have read in years. Here is just a selection of ‘oddities’, which the referees passed over:

1. The authors used an old set of criteria  (Oxford), which select a highly mixed population, dominated by people with psychological problems. They also tend to exclude patients with ME (since many have neurological signs suggestive of disease). The authors provide no information as to the number of subjects who had a post-infectious onset or what their main symptoms were (most apparently, did not have any of the symptoms in the list). All we know about them is that they had significant, ongoing, mental and physical fatigue which was not related to a medical condition, and that a substantial number thought that their symptoms did not have a ‘physical’ cause.

2. The researchers did not correct the HAD scores to give a more accurate estimate for depression (relevant for those with post-infectious CFS). They did not interview patients to establish current psychiatric diagnoses which may have influenced the results e.g. somatisation disorder.

3.    People attending a fracture clinic were used as controls for age and gender. People with MS and RA were rejected as they would already have been diagnosed at time of testing and knowing they had a medical disorder might have influenced their assessment of the listed symptoms.  Apparently, although the CFS patients had been ill on average 6 years, many had not been given a medical explanation, and often not even a diagnosis. This allowed the researchers to assess the patients’ views, before the input of a professional. However, one would not expect sensible people of normal intelligence to walk around for six years with a disabling illness without going to a doctor and asking (pressing) for a diagnosis. This is certainly not a behaviour commonly found in people with CFS (where help is sought fairly early on and diagnoses are offered, even if the patient doesn’t always agree with them). Another point to mention is that many acute-onset post infectious cases in the UK would have been diagnosed as having something like PVFS, and one can assume that many GPs would have ascribed their ongoing ill-health, at least, for the first couple of months, to a ‘medical’ condition. In other words, there is evidence in this report that the authors studied an unusual sample, and this should have been taken into account when discussing the results.

4.  The figures for the duration of the illness are also interesting for another reason. The mean for the patient group was 73.4 months with a range from 4 to 480 months. (The mean for the people attending the fracture clinic was 4.3 months, range .25-170.)  The inclusion of people with fatigue for four months suggests that some did not fulfil the Oxford criteria (which require a minimum of six months). The referees should have noted this, if nothing else.

5.  There is no information as to the number of patients who were asked to participate but who declined. The inclusion of response rates is common practice in medical journals, and they might have been illuminating here. 

6. The authors misrepresent the findings of other researchers (e.g. Sharpe et al) to emphasis the relationship between somatic attributions and poor outcome. Evidence to the contrary is not mentioned (Lawrie et al, Ray et al etc).

7. A sentence in the discussion suggests that the authors consider CFS to be a somatoform disorder. This ignores a lot of evidence to the contrary, not least, the general consensus amongst specialists that the term covers more than one condition. The literature as a whole does not support their view that patients with CFS tend to “view all symptoms as somatic in origin”.

From my perspective as an archivist, I can state that there is little support for the notion that attributions about the cause of CFS play a significant role in the perpetuation of fatigue. From my perspective as a scientist, I don’t think it’s valid to study patients using an out-of-date, flawed set of criteria, or to use data about symptoms they do not have in order to speculate about the possible effects of similar beliefs regarding symptoms, which they do have.  In my opinion, the quality of the design means that the results if this study are meaningless. If they are interested in the likely effects of certain beliefs, they should assess them and follow the patients over time. Then the correlations make sense.

A final point. The publication of this article is another example of the biased editorial policies operating in the UK, which accept papers from the King’s team, even though they are vague, unhelpful, and in some regards, scientifically poor. Using out of date criteria, selecting unusual patients (without explanation) and ignoring important findings which undermine one one’s personal views is not accepted practice in other medical specialities. We should not accept them in relation to CFS.

Given the politics, expect to read more about Butler et al than Bazelmans et al, van der Werf et al, De Becker et al or Ridsdale et al (Br J Gen Pract 2001, 51, 19-24.]


button - top