The studies found that Paxlovid, ivabradine, and computer games did not significantly improve symptoms of Long COVID. Further analysis that may help future research research has not yet been released.
Key points you should know:
- Three of the five initial Long COVID clinical trials from the National Institutes of Health (NIH) RECOVER program have started sharing results, all finding the interventions they tested did not significantly improve symptoms compared to placebos.
- The Paxlovid trial, RECOVER-VITAL, has yet to share any data from its analysis of viral persistence in participants’ blood, a highly anticipated component of this study.
- RECOVER-AUTONOMIC, a trial addressing dysautonomia, found the drug ivabradine alone did not help with symptoms but did lead to improvement when combined with lifestyle interventions.
- People with Long COVID are largely disappointed by RECOVER-NEURO, a trial addressing neurological symptoms, which tested computer games and psychotherapy rather than any pharmaceutical treatments.
- RECOVER-Treating Long COVID, or RECOVER-TLC, a newer phase of the program, promises to move faster and listen to community feedback better. Its new trials are set to launch this summer, two years after the initiative launched.
In summer 2024, Kelley Meister drove 90 minutes to the Mayo Clinic in Rochester, Minnesota to participate in the biggest Long COVID clinical trial launched in the U.S. at the time, part of the National Institutes of Health’s (NIH) flagship research program for the disease, called RECOVER.
Over eight site visits, Meister answered questionnaires, gave saliva and blood samples, and completed other testing. In the double-blind trial, some people with Long COVID received the COVID-19 antiviral Paxlovid, for either 15 or 25 days, while others received a placebo. When the trial was unblinded a year later, Meister learned that xe had received Paxlovid for 25 days.
Meister experienced some side effects during the trial, but ultimately found the drug helped with fatigue and post-exertional malaise (PEM). “There are other treatments that I think made more significant impacts on my health, but the Paxlovid study was what started to change the outcome from a downward spiral into a small and painfully slow, but still progressively upward movement back towards my previous healthy self,” xe wrote.
Nearly two years later, investigators from the trial have yet to publish a paper with their full findings. But early results indicate that the antiviral did not significantly improve participants’ symptoms compared to a placebo.
Two other trials from the RECOVER program have reported similar findings. A trial of the cardiac drug ivabradine, part of RECOVER-AUTONOMIC, a study intended to address dysautonomia in Long COVID, found that the drug lowered participants’ heart rates but did not meaningfully improve symptoms, investigators reported at a conference this spring.
And a controversial study to address neurological symptoms failed across its three treatment arms, the researchers shared in a JAMA Neurology paper.
Many people with Long COVID are frustrated, both by the delays in results and the interventions selected for testing. Finding treatments for Long COVID was one of RECOVER’s mandates when it launched in 2021 — the program’s name itself stands for “Researching COVID to Advance Recovery.” But five years on, some see little progress toward this goal.
“The first round of trials, for the most part, was just sub par,” said Hannah Davis, co-founder of the Patient-Led Research Collaborative and a patient representative who has advised RECOVER. “We deserved much better, much earlier.”
Most of these RECOVER trials tested behavioral interventions rather than pharmaceutical treatments that might actually address the underlying biological causes of Long COVID. People with Long COVID and outside experts have warned that one of the studies, which tests exercise as a treatment, may harm participants. People with the disease have also criticized a lack of transparency about how the tested treatments were selected as well as issues with patient engagement.
A newer arm of the program, called RECOVER-Treating Long COVID (TLC), aims to improve on this first round of trials with further studies that move faster, test more pharmaceutical interventions, and are better informed by community feedback. RECOVER-TLC is set to begin recruiting for two new studies this summer — nearly two years after the initiative’s launch in September 2024.
The delays in starting new trials and sharing findings are concerning to Jon Douglas, another patient-researcher who has advised RECOVER. Some people with Long COVID have been driven to try drugs off-label themselves rather than participating in clinical trials or waiting for their results, Douglas said — himself included. “I can’t wait. I don’t think a lot of patients can wait, either,” he said. “These trials should be able to go faster.”
The NIH did not respond to questions about when a paper from RECOVER-VITAL or other results from the first-round trials will be available.
Paxlovid results so far don’t include viral persistence analysis
Long COVID researchers and community members have closely followed RECOVER-VITAL, the program’s Paxlovid trial, since it was announced in summer 2023. While several clinical trials have evaluated the COVID-19 antiviral for Long COVID, VITAL is far larger than the other studies and includes a wealth of exploratory testing.
But the results of that exploratory testing — in particular, analysis of viral persistence markers in participants’ blood — are not yet available.
The first round of trials, for the most part, was just sub par… We deserved much better, much earlier.
Hannah Davis, co-founder of the Patient-Led Research Collaborative
VITAL recruited nearly 1,000 participants, just shy of its goal. Investigators assessed the participants in three groups based on their symptoms: one with cognitive dysfunction, one with autonomic dysfunction, and one with “exercise intolerance” or PEM.
In all three groups, participants who took Paxlovid (nirmatrelvir and ritonavir) did not see significant improvement in their symptoms compared to those who took placebos, according to results posted to clinicaltrials.gov in late March. This follows similar findings from Paxlovid trials at Yale and Stanford, as well as from a trial of a COVID-19 monoclonal antibody at the University of California, San Francisco.
Participants in RECOVER-VITAL improved at similar rates across the treatment and control groups. Outside commenters have wondered if the study may have used outcome measures that weren’t well suited to track improvements with the drug, or perhaps used a placebo that could also help. In the RECOVER trial, along with the Yale and Stanford studies, the control group took ritonavir, which is part of the Paxlovid regimen and has antiviral properties on its own.
Meister wrote that the trial’s questionnaires may have not fully captured hir symptom improvement: “I still experienced PEM throughout the trial (and continue to now), but I definitely have less PEM in terms of frequency and intensity than I did before the trial. But it was impossible to express that nuance through the questionnaires themself, and I feel that data was lost.”
Many people with Long COVID want to see Paxlovid and other antivirals tested over longer courses. “Antivirals tend to take a while to manifest results” for those with myalgic encephalomyelitis (ME), said Mark, a ME advocate who asked to use only his first name.
Another common challenge across these studies is the lack of a biomarker to identify which people with Long COVID may be best suited to benefit from antiviral treatments. People who have copies of the SARS-CoV-2 virus or its spike protein in their blood, for example, may be more likely to improve with a drug that stops the virus from replicating — but that doesn’t apply to everyone with Long COVID.
“For me, I see physical benefit when I’m on [Paxlovid],” said Robert DeRosa, who has Long COVID and has taken courses of the drug off-label, and is also involved as a patient representative for RECOVER.
DeRosa wants to see more detailed analysis of which study participants improve with Paxlovid and why. The current headline findings that “Paxlovid does nothing” for Long COVID have made it harder for him to access the drug himself, he said. Healthcare providers he sees “use these studies to guide their clinical practice,” and dismiss the evidence he shares from his own testing, he said.
Such analysis is part of the study design of RECOVER-VITAL. The study’s protocol document describes how participants may opt to share blood and saliva samples, which would be tested for “plasma spike protein antigen and other markers of viral persistence and reactivation” and compared before, during, and after participants took Paxlovid. Investigator Lindsey Baden also stressed the importance of this component of the study when presenting on the trial at the Long COVID International Conference in November 2025.
But no such findings are available in the preliminary results on clinicaltrials.gov. The trials at Stanford, Yale, and UCSF have shared findings that do evaluate different biological markers among study participants. Another Paxlovid trial, at the Karolinska Institute in Sweden, used participants’ samples to test a potential biomarker of viral persistence.
For all of these trials, the “most fascinating part” is analysis that can inform future research, Douglas said. “We’re learning that 25 days of Paxlovid is definitely not moving the needle. But we’re potentially seeing there might be a subset of folks who do have [some visible improvement] … And those parts of a publication will be really important.”
Ivabradine alone does not improve symptoms
When the NIH program announced its initial clinical trials in 2023, people with Long COVID were most interested in the Paxlovid trial and RECOVER-AUTONOMIC, a platform trial focused on addressing dysautonomia in Long COVID. The AUTONOMIC trial included two pharmaceuticals: ivabradine, an oral drug used to treat heart failure, and intravenous immunoglobulin (IVIG), an infusion of immune system antibodies from healthy donors. Both were tested along with behavioral interventions like wearing a compression belt and eating a high-salt diet.
Both of these pharmaceutical treatments are commonly used for postural orthostatic tachycardia syndrome (POTS) and other types of dysautonomia. Pam Taub, one of the trial’s principal investigators, proposed the ivabradine arm of the trial based on research she had done prepandemic “on ivabradine being helpful for patients with POTS,” she told The Sick Times.
Taub presented preliminary findings from the ivabradine study in late March at the American College of Cardiology’s annual conference. Unlike the Paxlovid results, this presentation came quickly on the heels of the study’s completion and analysis.
“We finished the trial in December, we got data in January, we put everything together for submission in January,” Taub said. “We got it out really, really quickly, mainly because of our desire to have results out [to the Long COVID community].” Slides from her talk are available online, and a paper with more detailed results is in progress.
In the presentation, Taub shared that ivabradine by itself did not significantly improve symptoms compared to the placebo, though it did lower participants’ heart rate. The study’s primary outcome measure was a symptom survey that Taub said has been used for dysautonomia drug approvals in the past.
However, participants who tried ivabradine along with behavioral interventions did see significant improvement. “The drug alone, in this very complex patient population, isn’t enough to translate into people saying, ‘I am better,’” on the questionnaire this study used, Taub said. But “when you combine [it] with lifestyle strategies which are evidence-based … there seems to be a benefit. And that’s what I see clinically.”
When you combine [ivabradine] with lifestyle strategies which are evidence-based … there seems to be a benefit. And that’s what I see clinically.
pam taub, researcher
Other providers have seen success with ivabradine clinically for POTS, and a recent review also found the drug lowered the heart rate of people with the syndrome. Still, the review concluded there was “limited evidence” for its effectiveness in improving symptoms.
Some experts are concerned that the “lifestyle strategies” in this trial include exercise, widely known to harm people with Long COVID who have post-exertional malaise or meet the diagnostic criteria for ME. Taub described the exercise program in the trial as “very tailored” to individual participants’ activity levels and designed with PEM as a consideration, as opposed to simply telling people to walk or run more. “A lot of the exercise has to be in a seated or lying position, and they gradually have to work up over time, to more of an upright exercise,” she said.
The protocol for the study includes measuring PEM with the DePaul Symptom Questionnaire after study visits, but researchers did not specifically assess it as a potential adverse effect of the lifestyle changes. No data on PEM are available in the preliminary results from Taub’s presentation.
Like the Paxlovid study, RECOVER-AUTONOMIC includes opportunities for exploratory analysis: samples for potential blood testing, autonomic function tests, data from wearable fitness trackers, and more. The researchers have “some plans” for more analysis, “but to really do more in depth, like biomarker analysis, we need more funding,” Taub said.
Patient-researchers who spoke to The Sick Times are interested in this analysis, hoping to see if there may be a subgroup of people with Long COVID who benefit from ivabradine. Davis hopes clinicians continue to prescribe it as “it’s obviously still helping a subset of the population. And I hope the final results will be able to highlight those subpopulations.”
As for the IVIG arm of the trial, data collection is still in progress and the researchers anticipate sharing results in late October or early November, investigator Peter Novak said during a webinar in March.
Neurological-focused trial did not test any pharmaceutical treatments
RECOVER-NEURO, which tested three treatments for neurological Long COVID symptoms, is the only one of the program’s first-round clinical trials so far to share results in a peer-reviewed scientific paper, published in JAMA Neurology in November 2025. But people with Long COVID are unimpressed by the trial, as it failed to study any pharmaceutical interventions.
The platform trial included three intervention arms: a computer program designed to help with cognitive function, called BrainHQ; a “cognitive rehabilitation” program, a type of psychotherapy, with individual and group sessions, paired with BrainHQ; and transcranial direct current stimulation (tDCS), in which a device sends low-level electricity to the brain, also paired with BrainHQ. For comparison, it included two control arms, testing a different set of computer puzzles designed as a placebo version of BrainHQ and a sham device designed as a placebo of tDCS.
“We were the first of the RECOVER trials out of the block in terms of recruiting, and ultimately we were the first to complete,” said David Knopman, one of the study’s principal investigators. All of the interventions could be done remotely, which helped with recruitment and retention, though participants did travel to study sites for four visits.
None of the interventions led to a significant improvement when compared to the control arms, based on the study’s primary outcome — a questionnaire called the Everyday Cognition scale (ECog), designed for people with dementia. Knopman noted this scale was not specifically validated in Long COVID prior to the trial, but other questionnaires that the study used to measure cognitive symptoms had similar results.
People with Long COVID were largely underwhelmed and unsurprised by these results when the paper was published. “In a shocking twist, we may need medication,” wrote Catherine Ozment, a New Mexico resident with Long COVID, in an email to The Sick Times.
Douglas said: “The NEURO [trial] is the most disappointing one, as a patient … I think many advocates did not want to see that trial even proceed.”
The NEURO [trial] is the most disappointing one, as a patient … I think many advocates did not want to see that trial even proceed.”
jon douglas, patient-advocate
The research team behind RECOVER-NEURO failed to listen to people with Long COVID, said Davis, who served as a patient representative on the study’s committee. “It was one of the worst experiences with patient engagement that I’ve had, doing it across hundreds of groups over six years,” she said. “I think the results go to show what happens when you both, a) don’t pick cutting-edge treatments, and b), have a clinical trial run by people who fundamentally are not knowledgeable about the space.”
Many experts and advocates have criticized the NIH for failing to put experts with more experience with infection-associated chronic conditions in charge of RECOVER’s initial work. With RECOVER-NEURO, one of the leading investigators “basically didn’t know what cognitive PEM was on the first meeting,” Davis said. “When you’re starting from that kind of position, of course the candidates are not going to be helpful.”
Knopman acknowledged that patient representatives had encouraged RECOVER to test more pharmaceutical interventions. However, he said that the evidence for the proposed drugs was “either nonexistent or underwhelming.” In his view, the study focused on treatment options that had some evidence behind them and would be safe to test in a phase 2 trial.
“In the world that I mostly work in, which is Alzheimer’s disease, it might take 10 years to come up with a drug that met the proper standards for moving to a phase 2 study,” Knopman said. “I think we chose from what we thought were the most promising of the rehabilitation options.”
More RECOVER trials in progress, with mixed reviews
Two more platform studies from the first round of RECOVER clinical trials have yet to share any results. One, RECOVER-SLEEP, aims to address sleep disturbances through testing the medication modafinil or the supplement melatonin, along with “light therapy.” The other, RECOVER-ENERGIZE, aims to address exercise intolerance through testing rehabilitation and address PEM through testing pacing.
RECOVER-ENERGIZE has been controversial since before it was even announced, with many advocates and researchers frustrated to see the U.S.’s flagship Long COVID research program testing an intervention that may harm participants. Despite calls for RECOVER to cancel this trial and test pharmaceuticals instead, the study has gone ahead, though with an extensive redesign. Namely, the exercise rehabilitation portion of the program states it will not include patients with PEM.
Laurie Gutmann, cochair of the clinical trials steering committee, told The Sick Times at last September’s RECOVER-TLC meeting that researchers working on the study “really still felt that this was an important treatment.” As in past statements from RECOVER leaders about the study, she failed to cite specific evidence supporting exercise as a Long COVID treatment.
Both RECOVER-ENERGIZE and RECOVER-SLEEP were set to complete enrollment by the end of 2025 and finish following up with participants by this summer, according to updates shared at the September meeting.
In another ongoing component of RECOVER, research teams are studying the disease’s underlying biology — many using samples and data collected from the program’s vast observational cohorts.
Meanwhile, new clinical trials through RECOVER-TLC are set to start this summer. This round includes three new studies: one testing low-dose naltrexone in children, one testing glucagon-like peptide-1 receptor agonist (GLP-1), and one testing the stellate ganglion block procedure. RECOVER-TLC also supported a preexisting trial that is testing the immune-modulating drug baricitinib.
The TLC initiative has promised to move faster and respond better to community input compared to earlier parts of RECOVER. Patient-researchers and advocates involved with the program have mixed reviews as to how that’s going so far.
“I do feel that they are trying to be more open and listening,” said DeRosa, one of the RECOVER patient representatives. For example, RECOVER-TLC has an online form where community members can propose treatments that they would like to see tested in trials. And the initiative has publicly shared draft protocols for all three of its new trials, allowing people to review and send comments.
But DeRosa is still frustrated by the time it’s taking for these trials to actually get started as well as concerns with transparency. “It’s really opaque” in terms of who is making decisions about these studies, he said.
In a recent webinar, RECOVER-TLC leadership announced that the initiative now has a community advisory board providing feedback on all trials — but did not share who is on that board, as they had not checked prior to the event whether members were comfortable having their names disclosed.
The TLC trials are “a dramatic improvement on the first round,” Davis said, “but in many areas are still very conservative. We’re not really at the point where we’re seeing the most experimental, the most promising, the most life-changing [interventions from RECOVER].”
Meister, the study participant in Minnesota, wrote of the delay in RECOVER-VITAL’s results, “I understand science takes time. However, it’s been six years since the start of the pandemic and with so few studies really looking at interventions, we need this information yesterday!”
Betsy Ladyzhets is a co-founder and the managing editor of The Sick Times.
All articles by The Sick Times are available for other outlets to republish free of charge. We request that you credit us and link back to our website
More research stories:
- RECOVER’s first round of clinical trials are failing. Will the next phase be better?
- New RECOVER-TLC trials won’t enroll until summer. Nearly two years after the program’s launch.
- RECOVER-TLC announces new Long COVID clinical trials, receives mixed reactions from patient community
- No “easy home runs”: Early Long COVID trials of Paxlovid and monoclonal antibodies failed, but the treatments still have potential
