Vaccination lowers the chance of developing long Covid, according to a large new study that also found that the risk of serious complications has diminished but not disappeared as new coronavirus variants emerged.
The study, published Wednesday in the New England Journal of Medicine, compared the health records of more than 440,000 Veterans Affairs patients who were infected with Covid-19 with records of more than 4 million uninfected people. The analysis found that cases of long Covid — also called PASC (post-acute sequelae of severe acute respiratory syndrome coronavirus 2) — fell among all participants during the Delta and Omicron eras of the pandemic, but dropped almost twice as much for vaccinated people when the Omicron variant dominated cases.
The new data about vaccination’s benefits come during a summer resurgence in Covid cases around the country, and as health officials prepare to roll out a vaccine updated against newer strains that Pfizer and Moderna said they could deliver in August.
“The study is helpful in validating the suspicion that the incidence of PASC decreased over time and in relationship to new variants (this question came up in my clinic just yesterday),” Hilary Goldberg, clinical director of pulmonary and critical care medicine at Brigham and Women’s Hospital in Boston, told STAT in an email. She was not involved in the study. “It is also helpful in validating that vaccination is protective in preventing the development of PASC, a question that has remained somewhat unsettled,” Goldberg said.
In the unvaccinated group, 10.42 out of 100 people early in the pandemic (before vaccines were available) had developed long Covid one year after being infected. In the Delta variant era (defined as June 19 through Dec. 18, 2021), 9.51 out of 100 unvaccinated people were diagnosed with long Covid, compared to 5.34 out of 100 vaccinated people. When the current Omicron era began (Dec. 19, 2021), the gap widened: 7.76 out of 100 unvaccinated people but only 3.5 out of 100 vaccinated people acquired long Covid.
“Vaccines very clearly work, but also clearly, they don’t totally wipe it out,” the study’s senior author, Ziyad Al-Aly, said in an interview. Chief of research at the VA St. Louis Health Care System, he is also a clinical epidemiologist at Washington University in St. Louis. “What we think is really important here is yes, long Covid has declined. But it’s not something that we can completely ignore.”
Earlier studies, including other research from the VA, have hinted at a protective effect for vaccination as well as the harmful impact of severe Covid infections and pre-existing medical conditions on chances of later long Covid. Jai Marathe, an assistant professor of infectious diseases at Boston University and director of the ReCOVer/ long COVID clinic at Boston Medical Center, points out that the new paper says the type or number of comorbidities did not seem to impact PASC.
“From a research perspective, there’s a strong impetus to understand the pathogenesis of PASC, given its constellation of symptoms,” he wrote to STAT. He did not participate in the NEJM paper. “If vaccination alters the immune response, we need to study this further to understand the interplay between viral infection and the immunological changes that contribute to PASC development.”
Al-Aly hesitates to call this the final word on the full extent to which vaccine coverage better protects people, urging more research. The observational study also describes how symptoms changed with the variants for unvaccinated people.
While debilitating fatigue was once a hallmark of long Covid, more recently gastrointestinal, metabolic, and to a lesser degree, musculoskeletal problems have overtaken fatigue among the unvaccinated, with GI and metabolic issues double the risk of the pre-Delta and Delta periods combined. The study, conducted from March 2020 through January 2022, measured how long Covid does its damage in 10 disease categories: cardiovascular, blood disorders, fatigue, gastrointestinal, kidney, mental health, metabolic, musculoskeletal, neurologic, and pulmonary.
Long Covid risk changed not just with the variants, but also with better hospital treatment, including antivirals, Al-Aly said. Vaccination may have lowered the amount of virus circulating and therefore the risk of long Covid compared to 2020, but that’s only a hypothesis. “Quantitatively, the risk changed dynamically over time, but also qualitatively. It suggests that each strain has its own fingerprints,” he said. “Long Covid is not monolithic.”
Michael Peluso, an infectious disease physician and an assistant professor of medicine at University of California, San Francisco, said he and other doctors may be seeing hints of the changing symptom mix. “It will make me more cognizant of GI and musculoskeletal symptoms in my assessment of long Covid,” Peluso, who was not part of the study, said in an email to STAT. “We certainly see these symptoms, but I would not say that they are the most prominent.
Goldberg wasn’t surprised by the symptom shift. “I find this reassuring in that we would expect the nature of the disease to evolve over time. If we did not see this, we could potentially question the robustness of the data in this analysis.”
To tease out the influence of vaccination versus variant, the study authors used a statistical method called decomposition analysis to conclude that 28% of the lower long Covid incidence in vaccinated people was due to changes in the variants and improved medical care, while attributing 72% of the improvement to the vaccines.
The study did not say how many vaccinations a subject received, Goldberg noted. “Gaps in the data remain — a common question from patients is what is the risk of reactivation or worsening of PASC symptoms from vaccination as compared to from infection,” she said. “We do not have sufficient information to address this question yet, and this question often informs patient decision-making.”
The study’s estimate of long Covid’s cumulative incidence is lower than a June report from the Agency for Healthcare Research and Quality published in JAMA last month that pegged long Covid’s reach at 7% of people who’ve ever had Covid-19. That’s higher than the NEJM paper’s 3% to 4% of adults in the Omicron era, but reflects the agency’s survey results as opposed to the NEJM study’s reliance on health records, Al-Aly said. Estimates of long Covid have swung wildly from early in the pandemic, before agreement was reached on just how to define the condition.
On Tuesday the Centers for Disease Control and Prevention said that people with disabilities face a higher risk of long Covid than people without disabilities. The data said 11% of people with disabilities had long Covid while only 7% of people without disabilities reported symptoms.
In other long Covid research this week, a study published Wednesday in Science Translational Medicine offered an explanation for damage to the lungs that leads to exhausting coughs, lung disease, and shortness of breath. Researchers from the University of Virginia showed that a signaling molecule named interferon-gamma stimulated an abnormal immune response, provoking inflammation and scarring in the lungs after Covid-19 infection. In mice, anti-interferon-gamma antibody treatment reduced inflammation and fibrosis after recovery from their acute infections. Drugs called JAK inhibitors, including one named baricitinib that was granted emergency use for acute Covid-19, “may merit consideration” for treating long Covid, the authors wrote.
Despite progress in research and declines in long Covid cases, the risk is still substantial, Al-Aly and others emphasized.
“Despite the risk of long Covid decreasing, it has not decreased to zero and we will still continue to see many cases of long Covid as SARS-CoV-2 continues to spread,” Peluso said. “This observation is really helpful for clinicians, researchers, and advocates to make the argument that more investment in clinical care and research for long Covid is going to be needed to meet the scope of the clinical need and to ultimately solve the problem.”
Declining uptake of vaccination is concerning, likely due to “Covid burnout,” Marathe said. “Living with a chronic condition like PASC can be debilitating. Therefore, educating and counseling patients and physicians to increase vaccine acceptance is a crucial public health task.”
Goldberg said patients with long Covid often ask her about the risks and benefits of updating their vaccination status. “This study will help to guide those of us caring for patients with PASC,” she said.
Peluso praised the paper’s value for answering questions patients ask “all the time” in clinic. “The risk with Omicron variants may be lower than it was with earlier variants, but it is still significant,” he said. “And being vaccinated is still the best thing someone can do, beyond avoiding Covid, to protect themselves against developing long Covid.”
STAT’s coverage of chronic health issues is supported by a grant from Bloomberg Philanthropies. Our financial supporters are not involved in any decisions about our journalism.