A Department of Veterans Affairs study found lower risk of COVID-associated cardiac events among vaccinated veterans, but critics say broader findings showed only modest protection and may have been affected by differences between vaccinated and unvaccinated groups.
By yourNEWS Media Newsroom
A new Department of Veterans Affairs study on COVID-19 vaccination and heart outcomes in veterans is drawing debate over how its findings have been presented, with critics arguing that headlines overstated the broader cardiovascular benefit.
COVID-19 vaccination rates now range from about one in 10 among children and pregnant women to about one in six among adults, according to Centers for Disease Control and Prevention estimates. The decline follows years of public messaging about the shots, which were portrayed as highly protective despite early recognition that they did not fully prevent infection or transmission.
The latest debate centers on a VA analysis published in JAMA Internal Medicine. The study examined more than one million veterans who received flu shots. About one-third of them also received a COVID-19 vaccine.
The most widely promoted finding was that veterans who received the COVID shot had a 37.7% lower risk of COVID-19-associated major adverse cardiac events, or MACE. JAMA Internal Medicine emphasized that result in a post on X. The study abstract also highlighted a 23.7% risk difference for all-cause MACE.
However, according to a Just the News report, critics noted that COVID-associated MACE made up only about 1% of all major adverse cardiac events in the study, limiting the practical size of the benefit. The study reported COVID vaccine effectiveness against all-cause MACE at 6.2%, with similarly modest findings for hospitalization and death at 6.6% and 7.1%.
The authors acknowledged that 10,000 vaccinations would be needed to prevent two COVID-associated MACE cases. They repeatedly described the benefit as “modest” and said the “absolute benefit” of COVID vaccination was “modest for the general cohort.”
Several media outlets emphasized the nearly 40% reduction in COVID-associated cardiac events while giving less attention to the study’s smaller all-cause findings and limitations.
Some critics also questioned whether the vaccinated and unvaccinated veterans were adequately comparable.
Epidemiologist Vinay Prasad, who returned to the University of California San Francisco this spring after serving as the Trump administration’s top vaccine regulator, reviewed the study’s supplemental tables and argued in his newsletter that the data “invalidates the entire paper.”
Because the research was observational rather than a randomized controlled trial, Prasad said the researchers needed to create two comparable groups whose main difference was COVID vaccination. He argued the supplemental data suggested that did not happen.
“All cause mortality curves separate by day 10. That is simply impossibly fast,” Prasad wrote. “The MACE separates by week 5, and the all cause death separates by week 2. The only explanation is that the no vaccine group is fundamentally different from the outset of the study.”
The Sensible Medicine newsletter, to which Prasad contributes, said the finding could reflect the “healthy vaccinee effect,” a known bias in which healthier people are more likely to seek vaccination. The newsletter called it a “JAMA IM editor & peer review fail.”
Prasad has previously written about medical interventions that appear to “work too fast” as a sign of selection bias. JAMA published one of his opinion pieces on the issue three years ago. The New England Journal of Medicine also published a letter from Prasad and former FDA colleague Tracy Beth Hoeg questioning whether a COVID booster study may have been affected by healthy vaccinee bias.
The Washington Post initially drew criticism for how it framed the VA study. An archived version of the article’s subheadline said the study found a nearly 40% reduction in “events like heart attack and stroke,” without specifying that the reduction applied to COVID-associated MACE rather than all such cardiac events.
The article body used more precise language, referring to MACE “linked to covid-19,” suggesting the error may have been introduced in the headline or editing process.
Former Surgeon General Jerome Adams, who served during Trump’s first term, shared the Post’s social media post about the study. The image attached to the post included the broader claim, while the text was more specific.
“Given the clear evidence linking COVID vaccination to better CV outcomes, discouraging or limiting access to vaccination seems difficult to reconcile with a goal of making America healthier,” Adams wrote in a post on X.
Vaccine Safety Research Foundation founder Steve Kirsch, formerly a major Democratic Party funder, criticized Adams and said the research was “too confounded to be relied up[on].”
The Washington Post later updated the article after former New York Times drug industry reporter Alex Berenson flagged the wording. Berenson noted that COVID-associated MACE represented roughly 1% of all-cause MACE in the study, or about five in 350 cases per 10,000.
“Five years after the first catastrophic failure of the mRNA jabs, legacy media outlets are still using falsehoods to push the shots on a population that neither wants nor needs them,” Berenson wrote.
Other outlets described the 37.7% finding more accurately but did not always emphasize how small COVID-associated MACE was as a share of total cardiac events. The Hill included the 6.2% vaccine effectiveness finding later in its story. STAT News did not include that figure.
The study’s corresponding author, Ziyad Al-Aly, did not respond to questions cited in the report about whether the study adequately matched vaccinated and unvaccinated groups or whether the headline finding should have been presented with more emphasis on the small proportion of COVID-associated MACE among all cardiac events.
The debate highlights an ongoing dispute over how observational vaccine studies should be interpreted and reported, particularly when relative risk reductions appear large but absolute benefits are small.
For the VA study, the key distinction is between a sizable reported reduction in COVID-associated cardiac events and a much smaller overall finding for major adverse cardiac events across the full study population.
Original article
Comments