ME/CFS and biased attention

Two posts here and here identify a recurring problem with the Wessely School theory of ME/CFS:

The complete lack of scientific evidence.

Here I consider the third strand of the theory:

H3: Biased attention causes, or exacerbates, the symptoms of ME/CFS and MUS.

According to this hypothesis, biased attention focusing towards symptoms, health-threats and related cues cause or maintain ME/CFS symptoms. I briefly review relevant studies.

Moss-Morris and Petrie (2003)

One of the first studies on the hypothesized attentional biases of pwCFS was by Rona Moss-Morris and Keith J. Petrie from the Department of Health Psychology, the University of Auckland, New Zealand (Moss-Morris & Petrie, 2003). They tested whether pwCFS have an “attentional information processing bias for illness-related information and a tendency to interpret ambiguous information in a somatic fashion”. Twenty-five CFS patients were compared to 24 healthy matched controls on a modified Stroop task and an ambiguous cues task in which they heard a tape-recorded list of 15 ambiguous illness words (e.g., vein/vain) and 15 unambiguous words. The participants were asked to write down the first word that came into their heads.

Moss-Morris and Petrie found no evidence for illness-related words creating greater attentional interference than neutral words on the Stroop task. However, on the ambiguous cues task, pwCFS made significantly more somatic interpretations than controls and this bias was associated with the extent to which they currently reported symptoms. Moss-Morris and Petrie concluded that :CFS patients have an interpretive bias for somatic information which may play a part in the maintenance of the disorder by heightening patients’ experience of physical symptoms and helping to maintain their negative illness schemas. Although patients did not show an attentional bias in this study, this may be related to the methodology employed” (p. 195). An interpretative bias is not an attentional bias, so this 2003 study produced a null result.

Hou, Moss-Morris, Bradley, Peveler and Mogg (2008)

investigated whether pwCFS show attentional bias towards health- threat information. On this occasion, the sample consisted of 14 pwCFS and 18 healthy controls. Hou et al. used a visual probe task which presented health-threat and neutral words and pictures for 500 ms and self-report questionnaires to assess CFS symptoms, depression, anxiety, and social desirability.  Compared to a control group, the CFS group showed an “enhanced attentional bias (AB) towards health-threat stimuli relative to neutral stimuli.” The finding of an enhanced AB towards health-threat information in pwCFS is claimed by the authors of being supportive of “models of CFS which underlie cognitive behavior therapy”.  However, it would be a long and dangerous leap from this modest result of a difference between groups (which may have a vast number of interpretations) to any kind of inference of causality.

Hou, Moss-Morris, et al. (2014)

continued to investigate attentional bias towards health-threat stimuli with enlarged samples of 27 CFS patients and 35 healthy controls. The participants did a Visual Probe Task to measure attentional bias, and an Attention Network Test measuring executive attention, alerting and orienting. They also completed self-report measures of CFS and mood symptoms. Compared to the control group, the authors state that the “CFS group showed greater attentional bias for health-threat words than pictures; and the CFS group was significantly impaired in executive attention. Furthermore, CFS individuals with poor executive attention showed greater attentional bias to health-threat related words, compared not only to controls but also to CFS individuals with good executive attention” (p. 9).

Hughes, Chalder, Hirsch and Moss-Morris (2016)

systematically reviewed experimental studies of attention and interpretation bias towards negative and illness-related information in people with CFS and healthy controls to December 2014. The results were overall inconclusive: “Some people with CFS have biases in the way they attend to and interpret somatic information. Such cognitive processing biases may maintain illness beliefs and symptoms in people with CFS” (italics are mine). When the stated conclusion is identical to the starting hypothesis, as in this case, it is safe to assume there were no new findings. Their review highlighted methodological issues in experimental designs.

The Nijmegen connection

Jointly with two colleagues in Nijmegen, Stephanie Nikolaus and Hans Knoop, Hughes et al. (2018) replicated a UK study with a Dutch CFS population. The authors claim that in two cultures, “people with CFS demonstrate biases in how somatic information is attended to and interpreted”. What has not been shown in either culture, however, is that these biases cause or exacerbate ME/CFS symptoms which is a fundamental claim in the Wessely School theory of ‘boom and bust’ ME/CFS.

Teodoro et al.’s (2018) systematic review

conducted a systematic review and suggested a general theory of what they term “functional cognitive disorder (FCD)” which is a disorder of cognitive dysfunction in the absence of an organic cause. They claim FCD is becoming increasingly prevalent and that the cognitive profiles in fibromyalgia (FM), chronic fatigue syndrome (CFS) and functional neurological disorders (FNDs) provide a ‘template’ for characterising their proposed new syndrome suggesting common underpinnings. They hypothesise that “pain, fatigue and excessive interoceptive monitoring produce a decrease in externally directed attention. This increases susceptibility to distraction and slows information processing, interfering with cognitive function, in particular multitasking. Routine cognitive processes are experienced as unduly effortful” (Teodoro, Edwards and Isaacs, 2018).

The results indicated that pwCFS do not show generalised abnormalities of attention or any general syndrome of a functional cognitive disorder. However, the review suggested to Teodoro et al. that some studies have shown that CFS patients may be prone to distraction in the Stroop task but this finding was not confirmed in all studies. Again, attentional bias to threat and towards emotionally negative information have been observed but unconfirmed. Owing to the very mixed bag of heterogeneous findings and “methodological shortcomings”, the authors were unable to make any general conclusions about the proposed new syndrome, or about CFS in particular.

Implications

1. To date, not a single published study confirms the hypothesis H3 that attentional bias causes or exacerbates symptoms of ME/CFS.
2. A mixed bag of heterogenous studies of mainly low quality containing multiple unresolved methodological issues. Two systematic reviews also yielded no definite conclusions about the association between attentional bias and ME/CFS.
3. As is the case for both H1 and H2, hypothesis H3, creates a chicken-egg problem: which comes first, attentional bias or the illness? Without controlled prospective studies, this question will remain unanswered, and the hypothesized causal link purely speculation.
4. This is the third hypothesis out of three in the Wessely School’s explanation of ME/CFS aetiology that receives no empirical support in the now extensive literature.
5. The theory, and its associated baggage of rag-tag treatments, should be abandoned.

Future posts will examine treatments recommended by the school, their effectiveness and safety.