CBT

By David Tuller, DrPH

Since the emergence of the phenomenon now called long Covid (or Long COVID, depending on news organization), skeptics have been out in full force. Even as huge numbers of people experience a range of sequelae after a bout of Covid-19, some experts maintain that common non-specific symptoms, like cognitive impairment and relapses of profound exhaustion, arise from anxiety, depression, post-traumatic stress disorder, hypochondriasis (renamed “health anxiety”) and other mental health issues. At times, their arguments have involved citing the PACE trial as supportive evidence of something or other.

It’s clear that anxiety, depression, post-traumatic stress disorder, hypochondriasis and other mental health issues can trigger and/or exacerbate any number of somatic sensations and associated fears. That possibility is not sufficient reason for the default assumption on the part of some that all of those whose prolonged symptoms currently resist satisfactory pathophysiological explanations are experiencing psychogenic or psychosomatic illness.

That has been the standard assumption about people with ME, CFS or ME/CFS for decades. Certainly the PACE trial and others have shown that cognitive behavior therapy (CBT) and graded exercise therapy (GET) are not effective treatments, despite continuing claims to the contrary. In issuing new ME/CFS guidelines in October, the UK’s National Institute for Health and Care Excellence (NICE) assessed the quality of evidence in support of these interventions as “very low” or merely “low.”

Not surprisingly, this decision drew protests from the CBT/GET advocates, who are eager to promote this same treatment approach for long Covid patients. In 2020, one of those fierce advocates, Professor Hans Knoop of the University of Amsterdam, received funding for a study of CBT for post-Covid-19 fatigue called ReCOVer: Can Cognitive Behavioral Therapy via the Internet prevent the fatigue symptoms of COVID-19 patients from becoming chronic? A controlled and randomized trial. The study follows the PACE model, which posits that some stressful event, often an infectious illness, serves as a trigger for fatigue and other symptoms, which are then perpetuated by “psychosocial” factors. I wrote about the proposed study here.

(I have previously criticized the work of Professor Knoop. In a commentary published alongside the 2011 PACE trial report in The Lancet, Professor Knoop and a co-author wrongly claimed that 30 percent of those who received CBT or graded exercise therapy met a “strict criterion” for recovery. I documented the commentary’s misrepresentations and misstatements here. It remains an embarrassment to the medical literature, as does PACE.)

Last month, Professor Knoop and his colleagues published the protocol for the study, titled “A randomised controlled trial testing the efficacy of Fit after COVID, a cognitive behavioural therapy targeting severe post-infectious fatigue following COVID-19 (ReCOVer): study protocol.” Here’s how the protocol explains the rationale for their approach:

“According to the cognitive behavioural model of post-infectious fatigue, the infection triggers fatigue and cognitive behavioural factors contribute to its perpetuation… These cognitive behavioural perpetuating factors can be addressed in cognitive behavioural therapy (CBT).”

And here’s how the protocol describes the Fit after COVID intervention:

“The main part of the intervention is based on an existing CBT manual for Chronic Fatigue Syndrome (CFS) by our research group and was adapted by experienced cognitive behavioural therapists… As the original CBT manual for CFS, Fit after COVID is based on a cognitive behavioural model of fatigue. According to this model, a disease or stressor (here: COVID-19) initially triggers fatigue while cognitive behavioural variables perpetuate fatigue. The seven perpetuating factors addressed in Fit after COVID are (1) disrupted sleep-wake pattern, (2) dysfunctional beliefs about fatigue, (3) low or unevenly distributed level of activity, (4) perceived low social support, (5) problems with processing the acute phase of COVID-19, (6) fears and worries regarding COVID-19, and (7) poor coping with pain.”

So the intervention addresses all the mechanisms through which patients, per the investigators, are bringing this fatigue on themselves. Neat! However, it is important to note that the protocol presents no evidence that the fatigue experienced by Covid-19 patients has anything to do with “dysfunctional beliefs about fatigue,” “perceived low social support,” “fears and worries regarding COVID-19,” or the other itemized factors. These speculations are apparently based on the beliefs of the researchers. Since they favorably cite results from the discredited PACE trial as part of the evidence base, their judgement can be questioned.

In the study, which is ongoing, more than 100 patients still suffering from fatigue months after Covid-19 are to be randomized to Fit after COVID or to “care as usual.” In other words, no effort has been made to structure the “care as usual” group in order to match–or control for–the experience of receiving an active intervention, which is likely to exert its own impact on recipients regardless of program content. It is therefore questionable whether this research should be granted the distinction of being called a “controlled” trial.  

Fit after COVID is a 17-week CBT program designed to be delivered mainly online, although users also have access to multiple conversations and consultations with an actual therapist. The primary outcome, measured at the end of treatment and six months later, is reported fatigue severity. Secondary outcomes include work and social adjustment, problems concentrating, and reported physical functioning.

This is an unblinded trial relying on self-reported, subjective outcomes–a study design that is a recipe for an unknown amount of bias. It is not surprising, for any number of psychological and emotional reasons, that participants who receive an intervention they are told could or should help them while engaging with a sympathetic therapist are more likely to report improvement than patients who receive neither the intervention nor the therapeutic contact. And yet objective measures might show no or minimal difference between the two groups, as happened in the PACE trial. Adherence to this study design is a deficiency in many ME/CFS trials, and the trend continues with this long covid one. The findings of such research are essentially uninterpretable.

The study does include one objective measure—participants will wear actigraphs, a wrist monitor that measure activity, for 14 days at baseline and 14 days at the end of treatment. In some earlier ME/CFS studies, actigraphs have been worn at baseline and during a follow-up period but not during the end-of-treatment period. At the follow-up assessment points, actigraph findings have not generally matched the self-reported improvements in fatigue and other measures. Some authors have delayed publication of these unattractive data while touting subjective findings as evidence of success.

For unexplained reasons, in this case Professor Knoop and his colleagues are not asking participants to wear actigraphs at the six-month follow-up period. Limiting this outcome measure to the assessment period at the end of treatment would appear to maximize the chances of positive results. Patients often report that they can increase their levels of activity through willpower for a short period–17 weeks, for example–before suffering a serious relapse. So actigraph readings right at the end of treatment are likely to be misleading.

Beyond all this, the study seems to treat the range of non-specific symptoms of long covid patients as if everything comes down to “fatigue.” It also seems to ignore the possible impacts or even the existence of the symptom of post-exertional malaise among this post-Covid-19 group. Although I have little hope that this study will yield useful findings, I assume it is likely to be presented and published as if it does. .