Trial By Error: Debate Over Recent Citation of Arguably Fraudulent PACE Trial

By David Tuller, DrP

*This post includes a correction, which is noted in the text.

Having a discussion about the PACE trial with someone who still maintains it was a well-conducted study is like debating the 2020 election with a Trumper who insists that Biden lost. These two distinct groups exhibit the same remarkable inability or willful refusal to understand and accept reality, whether it involves basic scientific constructs or presidential vote totals. After an initial exchange of perspectives, extended dialogue with either PACE or Trump defenders is pretty much pointless and serves only to provoke them to keep articulating their bogus arguments.

My initial 15,000-word investigation of PACE appeared in October, 2015—more than eight years ago. The PACE trial meets standard definitions of research misconduct and is arguably fraudulent, yet none of the PACE papers have been retracted. The authors stand by their findings, and so do the journals that published them–a true disgrace in the annals of science. Luckily, the trial’s reputation has been seriously damaged–and not only at UC Berkelely, where my epidemiology colleagues have deployed it in their graduate seminars as a case study in terrible research.

Bruce Levine, an emeritus professor of biostatistics at Columbia University, gave a lecture about PACE called “How NOT to Conduct a Randomized Clinical Trial.” The trial was harshly criticized in a letter to The Lancet signed by more than 100 experts and a host of international organizations (I wrote and organized the letter). The US Centers for Disease Control disappeared mentions of the PACE trial several years ago. In 2021, the UK’s National Institute for Health and Care Excellence found the PACE findings to be of poor quality and rejected them in developing its new ME/CFS guidelines, which reversed the agency’s prior recommendations of CBT and GET as curative treatments.

Despite all that, PACE continues to be cited favorably in the medical literature, as if it yielded actionable information about the benefits of psycho-behavioral interventions. That means its false claims continue to have an impact on patient care.

In early January, Nature Reviews Cardiology published a review article called “Cardiovascular autonomic dysfunction in post-COVID-19 syndrome: a major health-care burden.” This paper drew positive attention—in a tweet (or an X), well-known cardiologist and scientific commentator Eric Topol called it “best review I’ve seen for cardiovascular autonomy dysfunction, postural orthostatic tachycardia (POTS) in #LongCovid.” The lead author, Artur Fedorovski, is a professor of cardiology at Karolinska University Hospital in Stockholm.

Unfortunately, among the review’s list of 182 references—at #158—was the PACE trial. That citation was linked to a passage that quickly caught the attention of many observant readers: “Application of graded exercise therapy is especially important in the setting of coexistent ME/CFS to reduce the effects of the highly expected post-exertional malaise.” It goes without saying that the PACE trial most definitely did not document that graded exercise therapy can “reduce the effects of the highly expected post-exertional malaise.” This seems to be an odd interpretation of the PACE trial. Did Dr Fedorovski and his colleagues actually read it, or anything about the controversy surrounding it? Did they understand that GET is contra-indicated when patients have PEM? It seems not.

Todd Davenport, a professor of physical therapy at University of the Pacific in Stockton, California, countered Dr Topol’s praise, tweeting: “Any article published in 2024 citing the PACE Trial as evidence of safety and efficacy of graded exercise should not be taken seriously. It’s one citation, so maybe that seems unfair. But to neglect that level of scientific misconduct calls into question the authors’ judgement.” Much online discussion ensued, in that thread and elsewhere.

The journal has now published correspondence on this issue.

A letter from a group of Dutch, German, and Austrian investigators [*I originally left out “Austrian” in this sentence; I apologize for the oversight], joined by Davenport and David Putrino, a neuroscientist and physical therapist at Mt Sinai Health System in New York City, noted that “we cannot agree with the recommendations for graded exercise therapy for people living with long COVID who have post-exertional malaise.” In reference to the decision to cite the PACE trial, the letter-writers noted the following salient points: “The results of this trial have been called into question owing to substantial protocol deviations and retrospective adjustment of the criteria used to define recovery. A post hoc, per-protocol reanalysis of the trial data showed that the combination of cognitive behavioural therapy and graded exercise therapy was ineffective. Contemporary clinical guidelines for ME/CFS now advise against graded exercise therapy as a treatment and suggest just a supportive role for cognitive behavioural therapy.”

The response from Professor Fedorovski and his colleagues was unacceptable. They apologized for any “lack of clarity” in the initial article but then continued obfuscating. They suggested that their recommendations are consistent with the new NICE guidelines, which misrepresents what NICE concluded, as @_Lucibee noted in this comment on X. They also stated this: “Although a graded approach can have many meanings, we see that some authors use the term ‘graded exercise therapy’ to advocate a more standardized and less individualized approach to physiotherapy.” But they failed to acknowledge that this is exactly how PACE used the term. The PACE approach to GET, as outlined in the study’s owns descriptions, relied on regular incremental increases in activity and viewed physical setbacks as temporary effects of deconditioning and not as signs of potentially harmful pathophysiological processes. If the authors are not promoting the rigid PACE definition of GET, they had no business citing the trial in the first place.

Professor Fedorovski et al conclude their response with this: “The jury is still out; we should wait for the verdict and respect it when it comes.” No, the jury is not “still out” when it comes to the PACE trial. That’s bullshit. If these experts want to promote an individualized approach to exercise for some Long Covid patients, that’s one thing. But it is not appropriate for them to mis-cite the findings of the PACE study and the NICE guidelines to justify their position and then suggest that others have misinterpreted their intention. In this context, their “apology” about “lack of clarity” is essentially meaningless.