A COVID-19 vaccine effectiveness paper coauthored by CDC scientists and previously withheld from the agency’s Morbidity and Mortality Weekly Report has now been published by JAMA Network Open.
By yourNEWS Media Newsroom
A COVID-19 vaccine study coauthored by scientists from the Centers for Disease Control and Prevention has been published by JAMA Network Open after the agency’s acting director declined earlier this year to publish it in the CDC’s Morbidity and Mortality Weekly Report.
The paper, published on June 23, estimated that COVID-19 vaccination was 55% effective against hospitalization associated with COVID-19 and 50% effective against COVID-19-associated emergency department and urgent care visits.
“Adults can reduce their likelihood of severe COVID-19–associated outcomes by obtaining a 2025-2026 COVID-19 vaccination,” CDC epidemiologist Ruth Link-Gelles and coauthors wrote.
The CDC declined to comment.
Dr. Jay Bhattacharya, the acting CDC director, had previously declined to publish the study in Morbidity and Mortality Weekly Report, the CDC-run quasi-journal commonly used to release agency research.
“The study uses a test negative design that will not produce an unbiased estimate of efficacy (and it’s impossible to say which way the bias will go),” Bhattacharya said at the time.
The observational study examined patients who sought care across the Virtual SARS-CoV-2, Influenza, and Other respiratory viruses Network, a collaboration involving the CDC, data organization Westat and health systems including Kaiser Permanente.
Researchers reviewed patients who visited emergency departments, urgent care facilities or hospitals in the network from Sept. 3, 2025, to Dec. 31, 2025. They separated patients into two groups: those who tested positive for COVID-19 and those who tested negative.
Among 85,725 people who went to emergency departments or urgent care facilities in late 2025, 4.5% tested positive for COVID-19. Of those who tested positive, 206 people, or 5%, had received a COVID-19 vaccine in late 2025. By comparison, 12% of those who tested negative had recently received a COVID-19 vaccine.
Using that test-negative design, researchers estimated that COVID-19 vaccination provided 50% protection against COVID-19-associated emergency department and urgent care encounters.
The same method was used to estimate vaccine effectiveness against COVID-19-related hospitalization. Because a higher share of patients who tested negative had been vaccinated, researchers estimated vaccine effectiveness against hospitalization at 55%.
The study listed four limitations, including possible misclassification of patient vaccination status.
Several authors disclosed receiving grants from COVID-19 vaccine manufacturers, including Pfizer and Moderna. The study was funded by the CDC and Kaiser Foundation Hospitals.
In an accompanying editorial, Natalie Dean, an associate professor in the Department of Biostatistics and Bioinformatics at the Emory Rollins School of Public Health, wrote that the test-negative design has limitations but can still produce accurate vaccine effectiveness estimates.
Dean wrote that patients who test negative for COVID-19 “can approximate the underlying source population from which the COVID-19 cases arose, allowing investigators to estimate vaccine effectiveness without establishing a full denominator.”
She also noted that the method has been used for decades to estimate influenza vaccine effectiveness.
Dean was among those who defended test-negative studies during a CDC event convened by Bhattacharya to debate the research method.
Martin Kulldorff, an adviser to the Department of Health and Human Services, raised concerns during that event about using a test-negative control group in COVID-19 vaccine studies involving hospitalized patients with non-COVID-19 infections.
“It could be that they come from a more frail population. They’re more sensitive to infections and they know that and because of that they are recommended more strongly to get vaccinated,” Kulldorff said. “So it could be that those who come to the hospital for a different infectious disease, they are more frail and therefore they have a different probability of having been vaccinated. Has nothing to do with if the vaccine protects you against COVID or not. It just creates the bias. And the bias can go in in either direction.”
The publication places the previously blocked CDC-authored study into the broader scientific debate over COVID-19 vaccine effectiveness, observational research methods and how federal health agencies evaluate and release vaccine data.
Original article
Comments