By David Tuller, DrPH
In its current “Medical News in Brief” section, JAMA is touting and amplifying the questionable claims of a flawed trial to treat or prevent Long COVID published by one of the journals under its umbrella—JAMANetworkOpen. The JAMA headline: “Resistance Training Improves Long COVID Outcomes.”
Technically, the headline is true. In the trial, participants who received a program of resistance exercise reported statistically significant benefits on the primary outcome, a measure of physical capacity, compared to those who did not receive the intervention.
(The investigators also reported benefits on a few of many secondary outcomes. However, conducting multiple analyses increases the possibility of positive results from chance alone, and the investigators did not perform standard statistical tests to address that concern. That renders any claims about them questionable at best.)
The JAMA news-in-brief summary noted: “Intervention participants showed a greater overall improvement in distance traveled during the Incremental Shuttle Walk Test from baseline, an increase of 83 m vs 47 m in the control group after 3 months.” As I highlighted the other day in a post about the trial itself, there is less here than meets the eye.
The news-in-brief item, like the trial, doesn’t mention a key fact: The findings for the primary outcome, a standard measure called the Incremental Shuttle Walk Test (ISWT), were not clinically significant. The difference of 36.5 meters between the two arms was way below the 47 meters the investigators themselves had designated as the minimal clinically important difference (MCID) for the ISWT. As they described in the trial protocol, they used that value in determining the desired sample size, citing an authoritative 2022 analysis.
That wasn’t the only issue with the MCID in the trial. In my prior post, I forgot to mention another critical point: Whatever benefits participants in the intervention arm did or did not attain, the final results in both groups remained far below the average performance of healthy adults on the ISWT.
Overall, the participants achieved a mean distance of 389 meters at the end of the 12-month exercise program. In contrast, a 2013 study of “age-specific normal values for the ISWT” found a distance of 824 meters for 40- to 49-year-olds, 788 meters for 50- to 59-year-olds, 699 meters for 60- to 69-year-olds, and 633 meters for 70-year-olds and up. In other words, the participants in this trial remained seriously disabled across the board—another salient detail ignored by the investigators.
To hype statistically significant but minimalist results as if they demonstrated substantial improvement is obviously not helpful to patients. It is also deceptive. No one relying on the JAMANetworkOpen paper or the JAMA news-in-brief follow-up for clinical guidance would have any idea how poorly both arms did compared to healthy adults unless they took the time to dig a bit more deeply on their own.
Ignorng this information, like ignoring the fact that the difference between the two trial arms did not meet the threshold for clinical significance, is a form of misrepresentation. These methodological lapses reflect poorly on the investigators. Similarly, the journal’s decision to allow the paper to be published in this form reflects poorly on the integrity and competence of its editorial processes.
This isn’t the first time a Long COVID exercise trial has misrepresented the findings on the ISWT in this manner. In February, I wrote a post about a paper in European Respiratory Journal (ERJ), called “Post-Hospitalisation COVID-19 Rehabilitation (PHOSP-R): a randomised controlled trial of exercise-based rehabilitation,” in which something similar occurred. Participants in all arms of the trial—including the two intervention arms—performed way below normal levels for healthy people. In touting their findings, the investigators failed to address this point.
(Trudie Chalder was one of many co-authors of that study. As I wrote then: “It is a truth universally acknowledged (or at least universally acknowledged by smart researchers), that if the list of authors on an article includes Trudie Chalder, King’s College London’s mathematically and factually challenged professor of cognitive behavior therapy, then the article in question should most assuredly be expected to be short on, or utterly devoid of, intelligence and logical reasoning.”)
**********
Most Long COVID exercise trials ignore PEM, per Sick Times
A recent article in The Sick Times also addressed the issue with an article headlined “Less than 20% of Long COVID trials involving exercise even mention post-exertional malaise.”
The news organization analyzed the registration records for clinical trials and found that fewer than 20% assess or take into account post-exertional malaise, the defining characteristic of ME/CFS. Many people with Long COVID—the proportion is debated—experience PEM and quality for ME/CFS diagnoses. For these patients, exercise programs geared toward rehabilitation can be contra-indicated.
Here’s an excerpt from the piece, written by Toronto journalist Simon Spichak:
“According to an analysis conducted by The Sick Times, PEM is systematically neglected in research trials testing exercise interventions that could trigger it. Dozens of low-quality trials on exercise in Long COVID haven’t provided any answers and potentially harmed trial participants, and many never ended up publishing their results.
“The flawed idea central to these trials is that exercise is a panacea, and gradually increasing it could treat Long COVID, Jaime Seltzer, researcher and scientific director of the advocacy group #MEAction, told The Sick Times. As a result, the research won’t advance our understanding of the disease’s actual underlying mechanisms, said Seltzer, calling it “money down the drain.”
You can read the full article here.
