Your Questions About COVID-19 Breakthrough Infections, Answered

Here's what to know about infections that occur among people who've been fully vaccinated

A top down view of a person getting vaccinated by a nurse.
A healthcare worker administers a dose of the Pfizer-BioNTech COVID-19 vaccine Aug. 4 in Philadelphia.
Photo: Hannah Beier/Getty Images

When Paul Adamson, MD, first noticed he had a runny nose one day in mid-July, he figured it was an allergic reaction to the dust he’d kicked up doing yardwork. But when the runny nose didn’t clear up, he decided to get tested for COVID-19. Even though he was fully vaccinated, he had plans to fly to San Francisco to attend a baby shower, and he didn’t want to risk exposing the attendees—or the other passengers on the airplane.

“I didn’t actually think I had it,” says Adamson, an infectious disease physician at the David Geffen School of Medicine at UCLA in Los Angeles. But just to be safe, he bought a rapid antigen test at a grocery store. “It came back positive 15 minutes later.” He canceled the trip.

A follow-up PCR (polymerase chain reaction) test confirmed his COVID-19 diagnosis: Adamson had joined the growing ranks of fully vaccinated people with “breakthrough infections.” Such cases may become more common, with new evidence suggesting that vaccine-induced immunity wanes somewhat over time. On Aug. 18, U.S. health officials announced plans to offer booster shots to all Americans beginning in the fall.

Still, Adamson’s experience is the exception, not the rule. If you’re fully vaccinated against COVID-19, the chances you’ll contract the disease are much, much lower than they were before you got your shots. But they’re not zero.

Here, answers to questions you may have about breakthrough infections.

Do breakthrough infections mean the vaccines are no longer working?

No. Research has found that the existing vaccines are still highly effective at preventing hospitalization and death.

“What we really wanted was a vaccine to prevent severe disease,” says Kawsar Talaat, MD, an associate professor in the department of international health at the Johns Hopkins Bloomberg School of Public Health in Baltimore. “The fact that we got a vaccine that prevents most infections is totally a bonus.” Still, like all vaccines, the COVID-19 vaccines do not provide 100 percent protection—especially when a disease is spreading widely. 

It may be confusing to see reports like the one from a COVID-19 outbreak in Provincetown, Mass., which found that most people infected were vaccinated. (There were no deaths.) But think about seatbelts, suggests Josh Petrie, PhD, an epidemiologist with the University of Michigan School of Public Health in Ann Arbor. These days, most everyone wears one, which means that most traffic fatalities occur among people wearing seatbelts. But that doesn’t mean that seatbelts don’t significantly reduce traffic fatalities, just as COVID-19 cases among the vaccinated don’t indicate that the vaccines aren’t significantly reducing the number of infections. 

Real-world vaccine effectiveness estimates can be lower than in the clinical trials, which were conducted in a healthier group of people than the average population, says Gregory Poland, MD, director of the Vaccine Research Group at the Mayo Clinic in Rochester, Minn. Even so, a recent analysis published in the journal Immunity shows that the estimated real-world effectiveness of Pfizer and Moderna’s vaccines against severe disease has generally remained above 90 percent. Johnson & Johnson reported in July that its vaccine is 85 percent effective against severe disease. 

“I still have extreme confidence in the vaccine,” Adamson says. “Never did I feel scared that I was going to end up in the hospital or die from this. I think that I felt that way because of the vaccine.”

While the protection from the vaccines against hospitalizations and deaths has not yet fallen in the U.S., data from other countries, such as Israel, suggest that more breakthrough cases that lead to hospitalizations and deaths could begin to occur as vaccine efficacy wanes particularly among those who received their shots early on.

How common are breakthrough infections?

It’s not entirely clear, in part because this is not currently being tracked at the federal level.

In May, the Centers for Disease Control and Prevention stopped tracking all breakthrough infections. The agency now collects data only on breakthrough infections that lead to hospitalization or death—a small subset of the total.

“We don’t really have a routine surveillance system set up to catch all infections,” Adamson says. “Vaccinated people are less likely to have symptoms compared to unvaccinated people; therefore, they might be less likely to seek out testing.”

Still, some state-level data is available, and a recent analysis by the Kaiser Family Foundation offers some clues. Researchers found that in each of the 25 states that track this, less than 1 percent of vaccinated people have had a breakthrough infection. And a New York Times analysis of breakthrough infections that result in hospitalization or death found that in 40 states and the District of Columbia, vaccinated people represented between a fraction of a percent to 6 percent of hospitalizations and deaths from COVID-19. (Fully vaccinated people represent about half of all Americans.) 

What’s clear is that even when exposed to SARS-CoV-2, vaccinated people are much less likely to get infected than unvaccinated people. Adamson said his wife, who is vaccinated and was at home with him throughout his illness, never became infected herself.

Are certain vaccines more protective than others?

It’s not entirely clear. It’s true that in the clinical trials for the Johnson & Johnson vaccine, the measured effectiveness was about 66 percent, compared with the 95 and 94 percent effectiveness of the Pfizer and Moderna vaccines, respectively. But again, the data from clinical trials isn’t a true representation of real-world vaccine effectiveness.

More on COVID-19 Vaccines

“The trials were done in different times, in different places, with different virus variants, and different epidemiology,” says Talaat at Johns Hopkins. “Also, different definitions of what a breakthrough case was. You’re comparing apples to peaches.”

And the fact remains that all three of the vaccines currently approved under the FDA’s emergency use authorization in the U.S. protect very well against severe disease. That’s “the really, really important thing that I try to stress to my family and my patients and my friends that aren’t in the medical field,” says Crystal North, MD, a physician-researcher in the pulmonary and critical care division at Massachusetts General Hospital in Boston.

If I'm vaccinated and I get COVID-19, how sick will I be?

Most breakthrough infections are considered mild, and some are even asymptomatic. Still, that doesn’t rule out the possibility that you might run a fever and feel quite lousy for a few days, says William Schaffner, MD, a professor of infectious diseases at Vanderbilt University Medical Center in Nashville, Tenn.

That was Adamson’s experience. The next night after he developed a runny nose, he started to run a fever and had shaking chills and night sweats, followed the next day by muscle aches. The day after that, he was starting to feel better, but he lost his sense of smell. His wife had made him butter chicken, and as he was eating he realized it tasted “just like hot mush.” (His smell and taste gradually started to return after about two weeks.) Adamson says that while his experience would be considered a mild COVID-19 infection, it was still “the sickest I’ve been in over a decade.”

In a study published last month in the New England Journal of Medicine, among 1,497 fully vaccinated healthcare workers tested for COVID-19 at one Israeli medical center, 39 people tested positive for the disease. Of those, 13 people had no symptoms, while the others experienced symptoms such as congestion, aches and pains, loss of smell or taste, or fever. None were sick enough to be hospitalized.

Little research exists yet on long-term effects of a breakthrough infection, though some limited preliminary evidence suggests that breakthrough infections may be able to cause what’s known as “long COVID,” or chronic symptoms that last for weeks or months after the initial infection. In the NEJM study, 19 percent of those who had a breakthrough infection were still experiencing symptoms after six weeks. 

A very small proportion of breakthrough infections become severe. The overwhelming majority of hospitalizations and deaths from COVID-19 are among the unvaccinated.

If I’m vaccinated and I get a minor cough or a stuffy nose, should I get a COVID-19 test?

Yes. If things were different, and your community had low levels of COVID-19 circulating, and more elevated levels of other viruses such as flu and colds, testing wouldn’t be as important. This was the situation in many places earlier in the summer, Talaat says. But now that so much COVID-19 is being transmitted in so many communities, it’s best to err on the side of caution and get tested, even if you’re vaccinated, she says, and “even if it’s just the sniffles.”

What does the Delta variant mean for vaccinated people?

The clearest difference between Delta and other SARS-CoV-2 variants is Delta’s extraordinary transmissibility. “People who are infected with the Delta variant shed and produce so much more virus than the original parent strain,” Schaffner says. If you’re infected, and you breathe out, “think of it as a huge dark cloud of virus that you’re exhaling,” he says. 

Because Delta spreads so much more easily, it is causing a lot more people to get sick. And as more and more virus circulates in a given community, the chances rise that vaccinated people will get sick, too. “Numerically it’s going to look like more and more people who are vaccinated are developing disease,” Poland says. “That’s how the math works.” Again, that’s not a sign the vaccine isn’t working, but it does signal that not enough people are vaccinated to stop the spread of Delta.

Matters are made worse, Poland says, by communities lifting restrictions like crowd limitations and ending—or even outlawing—mask mandates. Many people are understandably tired of taking precautions, but relaxing mitigation measures gives Delta the opportunity to spread. “We’ve got the timber, we’ve got the gasoline, we’ve got the oxygen, and Delta’s the match,” Poland says. 

Vaccinated people without underlying conditions that impair the immune system can remain confident in the protections their inoculations provide against severe disease. But “waning vaccine-induced immunity over time may influence susceptibility to symptomatic breakthrough infections from the Delta variant,” North says. That’s one reason even vaccinated people shouldn’t completely abandon masks and other precautions.

Can vaccinated people spread COVID-19 to unvaccinated friends and family members, such as kids?

Yes. The extent to which this can occur is still unclear, Petrie says, but it can and has happened. The current thinking, he says, is that vaccinated people are probably less likely to transmit the virus than unvaccinated people, but that it’s still a real possibility.

And COVID-19 cases in children have been climbing in recent weeks, according to the American Academy of Pediatrics, accounting for about 15 percent of all cases reported during the first week of August. It’s not yet clear whether the Delta variant is more likely to infect and cause severe disease in children. It could simply be that because there are higher levels of disease all around, children—many of whom are not yet eligible for vaccination—become more likely to be infected. However, “the overall risk of severe illness is still much, much, much lower in children,” Petrie says.

Why did the CDC change its advice on masking for vaccinated people?

The CDC—after relaxing guidance on masks earlier this year and saying that vaccinated people can for the most part stop wearing them—now recommends that even vaccinated people don masks in indoor public spaces if there are high levels of virus spreading in the community. (That now applies to most communities across the U.S.)

The reason for this is the Delta variant, which is now the dominant variant in the U.S. Delta spreads particularly quickly among unvaccinated people, but it can spread among vaccinated people, too. The researchers who published the recent study on the Provincetown outbreak suggested that the spread of the virus among vaccinated people might be a sign that even in highly vaccinated communities, masking indoors in public would be wise, to further hinder the virus’s ability to spread. 

Continuing to use a mask when in public can be especially important for vaccinated people who live with unvaccinated children or people who are immune-compromised or otherwise at higher risk for disease from COVID-19. 


Catherine Roberts

As a science journalist, my goal is to empower consumers to make informed decisions about health products, practices, and treatments. I aim to investigate what works, what doesn't, and what may be causing actual harm when it comes to people's health. As a civilian, my passions include science fiction, running, Queens, and my cat. Follow me on Twitter: @catharob