Trial By Error: Just the Latest Gibberish from Professor Chalder

By David Tuller, DrPH

I’ve said it before and will undoubtedly say it again. Trudie Chalder, King’s College London’s mathematically and factually challenged professor of cognitive behavior therapy (CBT), is a one-trick pony. She writes what is essentially the same bad paper based on the same unfounded assumptions over and over again. Her apparent professional success represents, at least when it comes to this specific domain of inquiry, the broken state of scientific research, the triumph of mediocrity and incompetence, and the ethical bankruptcy of leading medical journals. From outside the culture of slavish deference to authority that seems to infuse British academia, Professor Chalder’s continuing ability to obtain major grant funding and publish endless muck is astonishing. Is there no mercy in this world for those of us forced by professional obligations to read this unceasing stream of sewage?

In Professor Chalder’s most recent study, she and her colleagues find an association between worsening fatigue and two of her favorite constructs– “all-or-nothing behavior” and “catastrophic thinking”–in patients with inflammatory bowel disease (IBD). The odds ratios are pretty tiny, however, at 1.13 for all-or-nothing behavior and 1.18 for catastrophic thinking—meaning that the odds of worsening fatigue are only marginally higher in those demonstrating these purportedly unhelpful behavioral or cognitive patterns. (The study was published by the journal Inflammatory Bowel Diseases.)

However banal and inconsequential, this sort of statistical finding is catnip for Professor Chalder. She seems never to have met an association she couldn’t try to spin as a causal relationship in order to justify the promotion of CBT as a solution. In this case, she seems to interpret this very minimal association to mean that these two identified patterns—all-or-nothing behavior and catastrophic thinking—are major factors in generating the reported worsening of fatigue.

This interpretation is implied, although not stated explicitly, in the headline: “All-or-Nothing Behavior and Catastrophic Thinking Predict Fatigue in Inflammatory Bowel Disease: A Prospective Cohort Study.” The word “predict” is doing a lot of work here. It creates the impression that these cognitive and behavioral patterns are to blame for the reported fatigue. But the association itself cannot be construed as evidence of that.

In reality, logic and common sense suggest that any causal relationships, however small, could easily run in the opposite direction than that presumed by Professor Chalder and her colleagues–that is, the fatigue itself is likely leading to the reported behavioral and cognitive patterns. Given fluctuating or worsening levels of fatigue, it makes sense that IBD patients would try to get as much done as possible when they felt well and less so when they felt worse. Moreover, if someone with pronounced or worsening levels of fatigue responds to questionnaires with a realistic appraisal of their condition, it could easily be interpreted by Professor Chalder and similarly biased investigators as “catastrophizing.”   

In this study, 167 patients filled out questionnaires at baseline and three months later. The primary outcome was Professor Chalder’s eponymous instrument, the Chalder Fatigue Scale. (Let’s put aside for now the various concerns that have been raised about the usefulness and accuracy of this questionnaire.) To assess possible “explanatory variables,” the investigators used an instrument called the Cognitive and Behavioural Responses to Symptoms Questionnaire (CBRQ). This instrument was designed and validated by experts, including Professor Chalder herself, who believe that fatigue and other “medically unexplained” symptoms should be treated with psychological and behavioral interventions, which they further believe are what they call “evidence-based.”

The CBRQ includes two behavioral subscales and five cognitive ones. The behavioral subscales are all-or-nothing behavior and excessive resting behavior. And here’s how the paper describes the cognitive ones:

“The cognitive subscales are (1) fear avoidance, which focuses on avoidance of exercise due to fear of worsening symptoms; (2) damage beliefs, which measure the belief that symptoms and their severity reflects true damage to the body; (3) catastrophizing about symptoms, which captures negative and inflated beliefs in anticipation of symptoms; (4) embarrassment avoidance, which quantifies avoidance of social situations due to feelings of shame about symptoms and concerns about others’ opinion of symptoms; and (5) symptom focusing, which assesses attentional focus towards symptoms.”

Each of these subscales only make sense if the underlying hypothesis of the CBRQ and this study—that these cognitive factors serve to perpetuate and exacerbate the fatigue—is valid. If the fatigue is driven instead by poorly understood pathophysiological dysfunctions connected to the primary diagnosis of IBD, then it would not be surprising for respondents to have higher scores on all of these subscales. These investigators, however, do not appear to acknowledge the possibility of such pathophysiological dysfunctions, so they view such straightforward accounts as “unhelpful” cognitions that need to be reversed through CBT.

Moreover, as one of the three lead investigators of the discredited and arguably fraudulent PACE trial, Professor Chalder still cites it in this paper as support for her claims, such as that CBT has been shown to be effective in producing “significant improvements” in fatigue in patients with what is referred to as “chronic fatigue syndrome.” (It is an insult to patients that even now Professor Chalder chooses to use this rejected name for the illness.) It goes without saying that PACE should never be cited as if the findings can be taken seriously, and no peer reviewers should allow such a citation to pass unchallenged.

As a secondary measure, the study included an instrument called the Work and Social Adjustment Scale (WSAS). As with fatigue, there was an association between worsening scores on the WSAS and the CBRQ subscales of all-or-nothing behavior and catastrophizing; the other behavioral subscale of excessive resting and the cognitive subscale of fear avoidance were similarly associated with worsening WSAS scores. The odds ratios in all four of these associations were as tiny as the associations involving fatigue. (Is it just me, or is there something confusing about having positive associations with both of CBRQ’s behavioral subscales–all-or-nothing behavior and excessive resting behavior? Don’t those two behavioral patterns kind of cancel each other out?)

In sum, Professor Chalder’s study of IBD patients yielded marginal associations between the primary and secondary outcomes and so-called “explanatory variables.” These findings would be negligible even if the investigators were correct in interpreting the relationships as causal in their favored direction and were not basing their interpretations upon their unproven hypotheses and questionable assumptions. The study’s inevitable conclusion–that “CBT interventions targeting all-or-nothing behavior and catastrophic thinking in IBD are warranted”–is self-serving bullshit.