A study on Covid vaccines that the Centers for Disease Control and Prevention’s acting director blocked from publication came out Tuesday in a different journal.

The findings show that Covid vaccines reduced the likelihood of severe illness by about half among adults last fall and winter. The study was originally scheduled to be released in March in the CDC’s flagship scientific publication, the Morbidity and Mortality Weekly Report (MMWR). Instead, it was published in JAMA Network Open, a highly regarded, peer-reviewed journal from the American Medical Association.

Acting Director Jay Bhattacharya raised concerns about the paper’s methodology after it had already undergone scientific review and MMWR editors had approved it, current and former CDC employees told NBC News in April.

At issue for Bhattacharya was the study’s “test-negative design” — an approach that compares the vaccination status of people who test positive for a particular disease (in this case, Covid) to the vaccination status of people who test negative.

The paper looked at adults who visited a hospital or urgent care with symptoms consistent with Covid across seven states from September to December last year. The group studied was tested for Covid around the time of their medical visit. Among those who tested positive and those who tested negative, the researchers calculated the odds of having received a 2025-26 formulation of the Covid vaccine. They found that the vaccine lowered the odds of a Covid-related visit to the ER or urgent care by 50% and of hospitalization for Covid by 55%.

Some public health experts saw Bhattacharya’s withdrawal of the study as political interference in the CDC’s scientific work or as an attempt to withhold evidence of vaccine safety. Health and Human Services Secretary Robert F. Kennedy Jr., who oversees federal health agencies including the CDC, has a long history of anti-vaccine activism. He referred to the Covid vaccine in 2021 as “the deadliest vaccine ever made.”

Bhattacharya told NBC News that he views longitudinal cohort studies, which follow large groups of people over a long period, as “a stronger design.” In the case of Covid vaccines, this type of study would track vaccinated and unvaccinated people over time, then compare the rates of illness. However, this type of research can be more expensive and time-consuming than test-negative studies, according to some experts.

In response to questions about concerns over the paper’s withdrawal, Emily Hilliard, a spokesperson for the Health and Human Services Department, said in a statement: “The CDC protects the public’s health by providing accurate, transparent, and trustworthy information. This requires applying the highest standards of scientific rigor, especially when findings may influence clinical decisions such as immunization.”

“Taking time to ensure analyses are methodologically sound and clearly communicated is always preferable to risking error,” she added.

In the newly published study, the authors write that a test-negative design is a “convenient and efficient method” to quickly evaluate vaccines in real-world settings. Outside researchers have said the approach comes with several advantages — it’s cost effective, provides timely results and compares groups with a similar likelihood of seeking out medical care.

However, the study has a few limitations. People who go to a hospital or urgent care when they’re sick may have more trust in the medical system, and could therefore be more likely to get vaccinated than the general population. And people in the study who tested negative for Covid had other respiratory illnesses, meaning they weren’t completely healthy.

In an op-ed in The Washington Post — which was first to report on the delay and subsequent cancellation of the MMWR study — Bhattacharya said the test-negative design disregards data about people who were never hospitalized. He also expressed concern that certain factors, like prior infections, could skew the results.

“The vaccine effectiveness estimates this method yields could be an overestimate or an underestimate; it’s impossible to tell,” he wrote.

The CDC recently hosted a public debate on the strengths and limitations of various ways to study vaccine effectiveness, including cohort studies and randomized controlled trials, which compare a group of vaccinated people to a group that got a placebo.

Natalie Dean, an associate professor of biostatistics and bioinformatics at Emory University’s Rollins School of Public Health, said researchers have been aware of the limitations of test-negative studies for decades but generally agree that they don’t undermine the conclusions.

“It just feels funny that this has become the difference between whether or not something makes it out in MMWR,” she said.

Dean, who participated in the peer review process for the JAMA Network Open study but was not involved in the research, said the quality of the study hasn’t changed since the CDC’s scientific review.

In an editorial in JAMA Network Open on Tuesday published alongside the study, Dean called the test-negative design “an important and practical approach” for studying vaccine effectiveness. Rather than switching to a different methodology, she said, test-negative studies could be made stronger by improving health records to include more detail about who is being tested and why.

“I get worried that it’s going to get caught up in some culture war, and it’s this major tool that we have,” she said in an interview.