Fact Check-NEJM study does not show that unvaccinated people with COVID are infectious for less time than the vaccinated

A widely shared Twitter post and the article to which it links misleadingly suggest that results of a small study published in the New England Journal of Medicine show vaccinated people remain contagious with a COVID-19 infection longer than unvaccinated people.

In fact, the study published online in late June by NEJM (here), found that the median duration of infectiousness (potential to pass on the virus) - as measured by the ability to grow virus in culture from nasal samples - was seven days among the unvaccinated, and six days among both the vaccinated and boosted groups, according to one of the study authors.

There were also no differences between groups in the proportion with a positive viral culture at 10 days, study co-author Mark Siedner, MD, of Massachusetts General Hospital in Boston, told Reuters. “In all three groups, less than 10% were still culture positive at that time,” he said by email.

“Our data do not suggest that vaccinated people recover more slowly from COVID-19,” Siedner added. “Quite the contrary: we found no significant difference by vaccination status (unvaccinated, vaccinated or boosted) in the time from a first positive test until PCR testing or viral cultures from nasal specimens became negative.”

The headline of an article published by, reading, “New England Journal of Medicine: Unvaccinated COVID Patients Are Contagious for LESS Time Than Those Vaxed or Boosted,” was posted by Aaron Kheriaty, MD (here) with a link to that article and the comment, “Gosh, I mean, who would have thought?” The tweet has been shared more than 14,000 times.

Twitter user responses to Kheriaty’s post include comments such as, “That means vaccinated have weaker immune system now…” (here) , “Wow finally some honest information” (here), and “What was the point of the jab again?” (here). Other users pointed out that the results and conclusions of the NEJM study referenced in the headline don’t support the headline’s assertion, including, “That isn't what the study was testing for nor the conclusion suggested by the doctors doing it” (here).


The article shared by Kheriaty on Twitter takes one piece of the NEJM study’s data out of context and “distorts it,” said study co-author Amy Barczak, MD, also affiliated with Mass General.

The article also highlights the wrong data to make its point. Viral culture, not PCR-positivity, gauges the potential for virus to be contagious.

In study graphs, the researchers chart the proportions and the actual numbers of individuals infected with either virus variant who remained PCR positive or viral-culture positive at 5, 10, and 15 days after their initial positive PCR test.

The researchers had recruited people newly diagnosed with COVID-19 by PCR test in the Mass General Brigham medical system between June 2021 and January 2022, and asked each person to provide multiple additional nasal swab samples, three times a week, for a total of two weeks. Each sample was tested by PCR for detectable virus, and cultured in the lab to see if the virus was potentially infectious.

PCR testing “measures whether there is any virus detected in the nasal swabs, irrespective of whether it is alive or dead,” Siedner explained. “Viral culture is a marker of contagiousness, with the presumption that the virus still needs to be alive to be able to be transmitted.”

Genome sequencing of the samples was also done to identify which virus variant an individual carried.

For their published analysis, the researchers included a total of 66 people. Of these, 34 people had been infected by the Delta variant and 32 by Omicron. At the time of their COVID infections, 16 people were unvaccinated, 37 had received the primary vaccine series, and 13 had also received a booster shot.

The study examined how many days it took from the initial positive PCR test, and from symptom onset, until people had a negative PCR test and until they had a negative virus culture.

On average, the study found, Omicron and Delta infections had similar durations. The median – meaning half the group took longer, half took less – time from the first positive PCR test until a negative PCR test was four days in the Delta group and five days in the Omicron group. The median time to a negative virus culture from whichever was earlier – symptom onset or the initial positive PCR test - was six days with Delta and eight days with Omicron.


The National Pulse article cites only data on the number of people PCR-positive at each time point to assert that “individuals who did not receive a COVID-19 vaccine were contagious for a shorter period of time.” The study graph shows, for example, that at day 10, one of the 16 unvaccinated people remained PCR-positive, as did eight of the 37 vaccinated people and one of the 13 boosted people. The National Pulse converts these numbers into percentages: “Regarding positive PCR tests, within the first 10 days of contracting the virus 68.75 percent of unvaccinated subjects were no longer contagious. In contrast, just 29.72 percent of vaccinated and 38.46 percent of boosted people were no longer contagious.”

The NEJM study authors point out in their report, however, that PCR tests cannot gauge infectiousness, which is the reason they used viral culture to determine whether live, potentially infectious virus was still present.

The National Pulse article then likens the NEJM study to other analyses that reach “similar conclusions that undercut the efficacy of COVID-19 vaccines,” and continues without citing evidence, “Natural immunity has also proven to be superior in fighting against the initial infection of COVID-19.”

“The overall goal of the study,” Barczak told Reuters by email, “was to understand how long people shed infectious virus (as estimated by culture positivity) after infection with delta or omicron variants. It was similar between the two variants, and took on average more than 5 days to convert to culture negative when timed from the earlier of initial symptoms or positive PCR, regardless of infecting strain or vaccination status,” she said.

When calculated by vaccination status, the unvaccinated group had a slightly longer median period of viral culture positivity indicating their potential infectiousness. “The median time of viral culture positivity in the three groups (unvaccinated, vaccinated, and boosting),” Siedner said, was seven days, six days and six days, respectively.

Even so, with a study population of just 66 people, the differences were too small to be statistically meaningful, the authors emphasize. “Importantly,” Siedner said, “all statistical estimates come with a range of uncertainty, based on the size of a study and other factors. When those are taken into account, there is no difference in the duration of shedding or PCR was similar in all groups.”


Misleading. Contrary to the claim, the study published by Barczak and colleagues in the New England Journal of Medicine found the median duration of infectiousness measured by cultured virus was one day longer among unvaccinated participants compared with those who had been vaccinated, but the size of the study group and the size of any differences between subgroups was too small for the differences to be statistically meaningful, according to the study’s authors. Claims based on PCR-positivity results alone also do not reflect infectiousness, and the study did not measure immune-system or vaccine responses at all, so no conclusions about vaccines or immunity can be drawn from it.