Whiting et al ⁽¹⁾ have made a valiant effort analysing the research on the management of chronic fatigue syndrome (CFS). However, were they totally impartial?

Reading the review, I noted that my own research had been given a 'validity score' of just 2. It was rated as 'poor' in five categories: baseline comparability of groups, follow-up, drop-outs,
appropriateness of controls and control for confounding. However, trials assessing cognitive-behaviour therapy (CBT) which had used the same test to control for confounding, had higher drop-out rates and notable differences in baseline scores, were rated as 'good'.

Using a quasi-experimental design suited to research within a clinical environment, I compared a well-defined sample with waiting list controls using valid measures, appropriate statistics and a six-months follow-up. Still, the review dismissed it as methodologically "very poor".

One factual error in the review was the failure to acknowledge the measures assessing fatigue, disability and activity. These should have been listed under physical outcomes but Table 2 only refers to the psychological variables (PS) and quality of life/health status (QOL).

The harshness of the authors' evaluation of my study contrasts sharply with that of another published around the same time ⁽²⁾. For example, their baseline comparability of groups was rated as 'good', despite the fact that the patients due to be given CBT spent twice as long in bed as the controls. As both groups has similar fatigue and disability scores, this anomaly is hard to explain. The randomisation had not only failed but the inclusion of more inactive patients in the treatment group could have
affected outcome.

In fact, the authors overlooked quite a lot of flaws in the CBT trials. For instance, the patients studied by Prins et al ⁽³⁾ were not required to fulfil all the listed criteria for CFS. As a result, they had also included people with idiopathic chronic fatigue ⁽⁴⁾.


Other inaccuracies included a reference for the London criteria written three years before they were formulated (I know this as I was a co-author), and a referral to diagnostic criteria for 'chronic fatigue
immunodeficiency syndrome' which do not exist and were not mentioned in the paper cited.

The accompanying editorial implied that this was an authoritative and reliable review ⁽⁵⁾. Perhaps an expert who had not spent the past five years extolling the virtues of CBT and graded exercise might have offered a slightly more measured appraisal?

Yours sincerely,
Ellen Goudsmit PhD
Chartered Health Psychologist.
London.