Chronic Denial

In April of 2020, physician-scientist Mady Hornig—then an associate professor at Columbia University—experienced an itchy throat, a swollen lymph node and a severely inflamed muscle radiating pain from deep in her ribcage. Three weeks after these initial symptoms, she developed a fever that persisted for two weeks. 

Before the pandemic, Hornig routinely worked 10- to 12-hour days. Trained in both immunology and psychiatry, she had long researched the role infection played in neuroimmune disorders, including the cluster of related diseases now widely called myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). But after Hornig recovered from the acute phase of illness, she continued to experience profound fatigue and could work only for short periods without resting. Sometimes she had difficulty initiating physical movements, such as typing or even speaking. And she frequently experienced a sudden, pounding heart rate—tachycardia—with no apparent cause. It felt like the battery that powered her body was truly shutting down. 

What Hornig did not have was a positive SARS-CoV-2 test. At the time she fell ill, it was still early in the Covid pandemic, and testing was limited. Moreover, her initial presentation, especially the delay in fever onset, was atypical, and standard clinical investigations revealed no abnormalities that could explain her prolonged and disabling symptoms. As a result, one of her long-time physicians said  she must be suffering from unconscious psychological issues and encouraged her to work at uncovering the root of these problems. 

“My doctor said, ‘Well, now we know that there must be some hidden unresolved stress or anxiety or problem,’ and I'm like, ‘No, I really don't think so,’” recalls Hornig, who is no stranger to reflexive psychological explanations for complex symptoms—especially in ME/CFS, which disproportionately affects women. Her doctor finally came around, acknowledging that her symptoms aligned with Long Covid rather than with anxiety or trauma. Hornig is now the president and co-founder of CORe Community (COVID Recovery through Community), a nonprofit focused on research into and treatment for Long Covid and related infection-associated conditions, especially among underserved populations.

As Hornig knows well from her work with ME/CFS, her doctor’s assumptions were grounded in a dismissive approach to chronic illness dating back more than 30 years. In the late 1980s, a powerful group of physicians and academic investigators, largely based in the United Kingdom, began promoting behavioral and psychological therapies as curative for an inexplicable disease. They rejected the name myalgic encephalomyelitis, which had been coined decades earlier for a similar presentation; instead, they called the disabling condition, most recently identified among patients in Lake Tahoe, Nevada in the mid-1980s, simply "chronic fatigue syndrome," or CFS. 

Hornig's doctor was being informed by the old view that patients with ME/CFS were severely out-of-shape, or deconditioned, in many cases because of a past, acute viral illness that had forced them to remain sedentary. The illness lingered in patients' minds, these experts opined, in the form of  “unhelpful” or “dysfunctional” beliefs about having, or continuing to have, an ongoing disease. In response, they recommended treatment focused on two interventions: a regimen of increasing activity called graded exercise therapy (GET) to address the purported deconditioning, and a specialized form of cognitive behavior therapy (CBT), designed to alleviate patients' allegedly misguided conviction that they were still sick.

That view of ME/CFS was ultimately abandoned by leading institutions setting the standard for medical care. Yet despite the update, old ways of thinking are still being applied  to patients like Hornig, who report long-lasting, often debilitating effects from Covid infection, stirring controversy again. 

To understand what's going on now with Long Covid, it's worth a deeper look at what happened with ME/CFS. One of the complications in treating ME/CFS has been a confounding core symptom that goes beyond profound fatigue: Patients experience a phenomenon known as post-exertional malaise, the delayed exacerbation of symptoms after even minimal exertion. Other key signs are cognitive dysfunction, disordered sleep, and orthostatic intolerance, the inability to stand upright without experiencing symptoms like dizziness, light-headedness, and nausea. 

For years, patients protested, to little avail, that the presence of post-exertional malaise meant that the favored treatments, in particular graded exercise therapy, were not only ineffective but potentially harmful. Finally, in 2015, the U.S. Institute of Medicine (now the National Academy of Medicine) published a major review of the scientific literature. "It is clear from the evidence," the group concluded, "that ME/CFS is a serious, chronic, complex, and multi-system disease that frequently and dramatically limits the activities of affected patients.” The report was one of a number of developments that dealt a significant blow to the prevailing psycho-behavioral paradigm, opening the door for further changes. 

In July 2017, after years of citing the contested research into graded exercise therapy and cognitive behavioral therapy for ME/CFS, the US Centers for Disease Control and Prevention (CDC) finally, quietly dropped the recommendations for these treatments. The decision was first reported on Virology Blog followed by a report in STAT News ((I wrote the first and co-wrote the second) and confirmed by the ME Association, which noted that the CDC had removed CBT and GET from its website without fanfare, following sustained patient and expert critique.

In October 2021, the UK’s National Institute for Health and Care Excellence, or NICE, which issues clinical guidelines for many conditions, recommended against the use of graded exercise therapy and advised that CBT should not be offered as a cure but merely to support patients in managing their symptoms. In developing its updated ME/CFS guidelines, NICE found the evidence for these treatments was of "low" or “very low” quality. The full NICE guidelines are available here. This appraisal of evidence quality is discussed further in a peer-reviewed summary.

Now, despite the broad discrediting of such treatment strategies for ME/CFS, the same narrative is resurfacing. Some of the same investigators, academics and clinicians in the field have begun to promote the same old theories and strategies for the treatment of Long Covid—an umbrella term for anyone experiencing prolonged symptoms after Covid-19, although researchers are still trying to refine their definition of the disease. Many Long Covid patients report symptoms closely resembling those of ME/CFS, reinforcing the long-standing awareness that viral illnesses cause chronic disability in a small proportion of those who get sick. It seems clear that ME/CFS and these manifestations of Long Covid are interrelated. Rigorous research into either condition could illuminate mechanisms behind both. But misguided interventions once again risk real harm to real patients.

And that's what seems to be happening here. Notably, the ReCOVer study (formally titled “Efficacy of Cognitive-Behavioral Therapy Targeting Severe Fatigue Following COVID-19: Results of a Randomized Controlled Trial,” published in Clinical Infectious Diseases,  on April 15, 2023), relied on the  psychosomatic theory to explain the ongoing sickness of patients like Hornig, recommending cognitive behavioral therapy as a viable treatment. The senior author, Hans Knoop, is a professor of medical psychology at Amsterdam University Medical Center. He has spent decades investigating and promoting psycho-behavioral treatments such as cognitive behavioral therapy for ME/CFS and related illnesses.  

One key issue with the ReCOVer study is its rejection of its own objective data. While conducting this randomized controlled trial, the researchers deployed actimeters—wrist-worn devices that objectively tracked physical activity before and after treatment. Though not mentioned in the published paper, the authors subsequently acknowledged that actigraphy data showed no measurable benefit in activity levels in the group that received the intervention. Given the obvious link between fatigue and physical function, these null findings should have carried weight. Instead, the authors dismissed them, claiming that subjective reports of fatigue are unrelated to actual movement—an argument that undercuts the logic of the cognitive therapy itself. If physical activity isn't relevant, then why measure it at all?

The study has other flaws as well. For example, the study was unblinded, meaning both participants and researchers knew who was receiving the intervention. Essentially, participants were asked if they felt better after having received an intervention that they knew was supposed to make them feel better. The 114 participants were all suffering from what was identified as severe fatigue three to 12 months after their coronavirus infections. They were randomly placed either into a group receiving the cognitive behavioral therapy program, plus care as usual, or a comparison group receiving only the latter. 

The CBT program, which ran for up to 17 weeks, was specifically designed to target factors believed to be perpetuating the symptoms, including “unhelpful beliefs” about fatigue, “fears and worries regarding Covid-19,” and “problems with psychological processing of Covid-19,” according to the published trial materials. Given the study design, the results are inherently vulnerable to bias: Participants might report feeling better not because the treatment worked, but because they hope or expect it to.

Unlike with pharmaceuticals, it is difficult if not impossible to blind participants to psychotherapeutic or behavioral interventions. That makes it challenging to design trials that truly can measure the impact of non-drug approaches. Recognizing this limitation, investigators in the field have, over the years, tried a range of objective measures of function, including the level of physical activity over the course of several days or a week, as measured by worn devices like actimeters. However, these and other objective measures (such as how far people can walk in six minutes, or their employment status) have almost invariably failed to match subjective reports of benefit. 

If participants in a trial know they are receiving an active treatment, and they know that this treatment has been promoted as successful, they would be far more likely to report improvements after the treatment than people who received no such treatment at all. The trial was therefore almost guaranteed to produce at least modestly positive findings. Assertions that these findings represent the real impact of the therapy therefore cannot be taken at face value.

The bottom line is this: The psychosomatic interpretation of Long Covid recycles a discredited framework, one that has repeatedly failed patients with ME/CFS. Given the poor track record of these methods in ME/CFS studies—where reported subjective benefits were not matched by objective findings—it is not surprising that Long Covid studies relying on the same paradigm would face similar shortcomings. More dispiriting still, the peer-review system, which is supposed to act as a guardrail, appears inadequate to protect against dubious psychosomatic conclusions reappearing in research publications. Publication in a high-profile journal unfortunately does not guarantee the findings are reliable or free from bias.

A patient like Mady Hornig, with her deep medical training and expertise, is capable of assessing complex studies herself and confidently rebutting her own doctor's views. But Hornig notes that most ME/CFS and Long Covid patients would find it difficult to rebut seemingly authoritative pronouncements from their health care providers. That's why she and other experts are continually pushing for better physician education and greater funding for research that moves beyond psychogenic assumptions. Although no cure exists for either condition, some biological treatments are currently under investigation. At this stage, what is most urgently needed is more rigorous research, along with significantly more compassion.

“If a doctor who has known me for a long time could suggest that post-Covid tachycardia and other symptoms likely reflected some deep-seated unconscious neuroses, then how would anyone without a medical background manage to be properly heard and guided through this?” Hornig asks. This time, patients and researchers like Hornig are determined not to let history repeat itself unchallenged.