More than 50 million Americans have now gotten their first shot against Covid-19, and 25 million have gotten two shots. There’s strong evidence that the two Covid-19 vaccines Americans are getting — one by Pfizer/BioNTech, the other by Moderna — are highly effective at preventing illness, hospitalization, and death.
Despite that fact, public health officials and media outlets have been warning that vaccinated people need to behave largely how they did before they were vaccinated. That’s because we don’t know as much about the vaccines’ effectiveness at preventing transmission to others. A vaccinated person may be well-protected from Covid-19, but if they carry the virus, could they possibly infect the people around them?
But a growing body of evidence suggests the Pfizer/BioNTech and Moderna vaccines do, in fact, cut down on viral transmission. Two recent studies show some pretty favorable results — one from the UK that found that two doses of the Pfizer/BioNTech vaccine cut down by 86 percent someone’s chances of developing an infection that they could pass along, the other a study in Israel that found an 89.4 percent reduction (though it should be noted that the Israeli study has yet to be fully released). These findings are consistent with what we know about vaccines and transmission in general.
In other words, even as we wait for more definitive studies on the vaccines’ effects on transmission, more and more scientists think we do have enough information to feel pretty good about the vaccines’ capacity to give us back a semblance of normalcy as we approach a year of life in a pandemic.
In an opinion piece, Johns Hopkins epidemiologists M. Kate Grabowski and Justin Lessler argued, “We are confident vaccination against COVID-19 reduces the chances of transmitting the virus.”
“I have been very cautious due to limited evidence on transmission effects but agree with [Grabowski and Lessler] that a large transmission effect is the best explanation of the limited evidence to date,” Harvard School of Public Health epidemiologist Marc Lipsitch said in February.
In a new preprint, he and Harvard epidemiologist Rebecca Kahn laid out some principles for how to evaluate data from observational studies like the UK and Israel ones. Applying their principles to Moderna data (more on this below), they estimate that “one dose of vaccine reduces the potential for transmission by at least 61%, possibly considerably more.”
Even Dr. Anthony Fauci has sounded an optimistic note. “The looming question is, if the person who’s been vaccinated gets infected, does that person have the capability to transmit it to another person? Some studies are pointing in a very favorable direction,” he said in a White House briefing in mid-February.
Studying exactly how much a vaccine affects transmission is very difficult. It requires exceptionally good contact tracing, which few countries have, or inference from lots of different forms of limited evidence. Uncertainty remains about exactly how much the vaccines reduce transmission — and that uncertainty has led many public health officials to be cautious in their public statements.
But that cautiousness can end up misleading the public, giving people the impression that scientists have no information at all. That, in turn, could also lead to vaccine hesitancy. Some people may think, if I get vaccinated but I still have to continue masking and social distancing at all times, then why get vaccinated at all?
“In their own lives, medical experts — and, again, journalists — tend to be cleareyed about the vaccines. Many are getting shots as soon as they’re offered one. They are urging their family and friends to do the same,” David Leonhardt argues in a New York Times piece. “But when they speak to a national audience, they deliver a message that comes off very differently. It is dominated by talk of risks, uncertainties, caveats and possible problems. It feeds pre-existing anti-vaccine misinformation and anxiety.”
The vaccines do reduce transmission. They do take us a big step closer toward life beyond the pandemic. And the messaging from our institutions should start reflecting that.
The two new studies looking at the Pfizer/BioNTech vaccine, explained
Now that many people have been vaccinated against Covid-19 — about 50 million in the US — new research is coming out every day that clarifies the transmission picture. Two key new studies look at the rate of disease among people who got two doses of the Pfizer/BioNTech mRNA vaccine. That’s an important figure for judging how effective vaccines are for transmission — the lower the rate of infection, the lower the rate of transmission.
A little step back here to explain where things stand on vaccines: There are several vaccines approved in different parts of the world, with various approaches and levels of effectiveness. The Moderna and Pfizer/BioNTech vaccines were the first to be approved for use in the US, and there’s more data available about them, and to keep the scope of this piece manageable, it will focus on just those two vaccines.
A working paper recently published with The Lancet’s preprint publication program looked at health care workers in the United Kingdom who were vaccinated with the Pfizer/BioNTech vaccine. It finds that the vaccine doesn’t just make people less likely to get symptomatic infections (which we already knew from Pfizer/BioNTech’s initial trials) — it also makes them much much less likely to get infected at all. “Vaccine effectiveness was 72% ... 21 days after first dose and 86% ... 7 days after the second dose,” the study concludes.
That’s lower than the 95 percent headline number you might have seen, but that 95 percent measures symptomatic infections; this measures all infections, even “invisible” asymptomatic ones, through routine testing of healthy people.
Another recent paper, this time out of Israel, looked at the Pfizer/BioNTech vaccine as well. (A big caveat: The findings were announced in a press release, but the study itself hasn’t been released yet.) It found a drop of 89.4 percent in infections among people who got two doses of the vaccine, compared to unvaccinated people. We don’t have as much information from this research as we’d like yet, as the paper has yet to be made public, and given the methodological challenges of estimating transmission, the details of the paper matter a lot. But that number is similar to the one from the UK study.
Even though the studies focused only on the Pfizer/BioNTech vaccine, there’s reason to believe that the results translate to the Moderna vaccine as well. The two vaccines work very similarly. Both contain a set of instructions to the RNA in our cells to build a protein very similar to the “spike protein” in the coronavirus. Then the immune system notices the intruder and responds, producing antibodies that’ll protect against the coronavirus later.
Because the two vaccines work very similarly, the researchers I spoke to said it was overwhelmingly likely that they both block transmission to a similar degree. As a result, we can assume — though with some uncertainty — that evidence of strong infection reductions from the Pfizer/BioNTech vaccine also likely applies to Moderna.
But even before the most recent research came out, we already knew that the vaccines would help curb transmission. For one thing, the Moderna and Pfizer/BioNTech mRNA vaccines reduce the chances of getting a symptomatic case of Covid-19 by 94 percent and 95 percent, respectively. That is a promising starting point — if a person doesn’t get Covid-19, then they can’t pass it on.
But what about asymptomatic cases?
In their initial clinical trials, Moderna and Pfizer didn’t study whether vaccinated people got asymptomatic cases of Covid-19 — that is, people who tested positive for the coronavirus but did not suffer any symptoms. However, when people went in for their second shot, Moderna did give them a nasal swab test for Covid-19. In a supplement to its submission to the FDA, Moderna says that 14 of the 14,134 vaccinated people had Covid-19 (with no symptoms at the time) and 38 of the 14,073 people in the control group had Covid-19 (with no symptoms at the time).
That rules out one big worry about the vaccines: that they might make Covid-19 mild in vaccinated people — so mild they don’t experience any symptoms — without actually preventing it. Instead, it was clear from back in December that the vaccines reduce asymptomatic infection as well as reducing symptomatic infection.
Using Moderna’s nasal swab test data, infectious disease biologist Marm Kilpatrick at UCSC estimated that the vaccine, after a single shot, reduces a person’s odds of infection with Covid-19 by up to 90 percent. (When I emailed him, we determined that with some more pessimistic assumptions, the reduction might be more like 78 to 88 percent.)
Lipsitch and Kahn took a different approach, and estimated from the same data that “one dose of vaccine reduces the potential for transmission by at least 61%, possibly considerably more.” Of course, the overall efficacy of the vaccine after both doses will almost certainly be higher.
The recent data on the Pfizer/BioNTech vaccine in Israel and in the UK backs up that idea. It suggests that after two shots, the vaccine is 85 to 90 percent effective at preventing infection with Covid-19.
There are some caveats. The data from the UK and from Israel comes from observational studies, not randomized controlled trials: If the people who’d been vaccinated differ from people who haven’t, the study’s assumptions might not hold. Researchers do their best to adjust for this, but any adjustment will be imperfect. In addition, getting the vaccine could change behavior — the vaccinated might take more risks, and they might be less likely to seek Covid-19 testing or be required to provide negative test results.
So this estimate shouldn’t be considered definitive. But it lines up with other sources of evidence, and it suggests that, overall, the vaccine is likely highly effective — in the 80 to 90 percent range — at preventing infections. And low infection rates mean low transmission rates.
Viral load and reduced odds of transmission
But let’s say a person who has been vaccinated still gets infected with Covid-19. That’s not great, but the vaccine likely continues to protect the people around them, according to the research so far. That’s because of another consideration: viral load — that is, how much virus can be measured in a patient’s nose and throat.
Not everyone who has Covid-19 is equally likely to transmit it. A study recently published in The Lancet based on research from contact tracing in Spain has found a very strong association between viral load and how many other people the patient infects, as well as how serious the infections in other people are.
This isn’t very surprising. Viral load determines how much virus you are coughing or breathing into the air, which determines whether other people get sick. And if they get sick with an unusually large dose of the virus, it’ll have a “head start” at infecting them, and they’re likely to get sicker.
“In our study, the viral load of index cases was a leading driver of SARS-CoV-2 transmission. The risk of symptomatic COVID-19 was strongly associated with the viral load of contacts at baseline,” the study concludes.
The impact of the vaccine on transmission, then, will be the product of two factors, co-author Michael Marks, an epidemiologist at the London School of Hygiene and Tropical Medicine, told me: lower chance of getting infected, and lower viral load if infected.
We already covered the former point above; what about the latter? Do the vaccines cut viral load?
On this front, there’s great news in another preprint based on data from Israel: The Pfizer/BioNTech mRNA vaccine appears to cut viral load dramatically, so people who do get Covid-19 after the vaccine have less of the virus in their nose and throat, making them less likely to infect other people.
“We find that the viral load is reduced 4-fold for infections occurring 12-28 days after the first dose of vaccine. These reduced viral loads hint to lower infectiousness, further contributing to vaccine impact on virus spread,” the study concludes. This research is just a preprint, not yet peer-reviewed, but if the data holds up, it would suggest that vaccinated people who test positive and are infectious are still significantly less infectious than unvaccinated people.
“The data is certainly intriguing and suggestive that vaccination may reduce the infectiousness of COVID-19 cases, even if it does not prevent infection altogether,” Virginia Pitzer, an infectious diseases modeler at the Yale School of Public Health, told Nature.
Many of the caveats discussed above apply to this study, too. This research from Israel is an observational study, not a randomized controlled trial. However, the vaccinated people had the same average viral load during the first 12 days after vaccination as the unvaccinated people, and only after 12 days did a difference start to emerge, which suggests the vaccine is what is producing the difference.
In total, vaccination unambiguously makes people less likely to get a case of Covid-19. Then, if a vaccinated person does get a Covid-19 case, preliminary Pfizer data from Israel suggests they’ll have lower viral loads, which other research has established makes them less likely to pass on the virus. And because of the lower viral load, if they do infect another person, the infection is less likely to be serious.
To be clear, the transmission point is based on early data — there’s still uncertainty about how exactly lower viral loads in vaccinated people will translate to lower infectiousness. But “some data” is different from “no data.”
How we should and shouldn’t talk about uncertainty
There isn’t significant doubt among epidemiologists that vaccines somewhat cut transmission.
First, almost all vaccines do that, so it was a good starting assumption before we had any data at all. (There are a few exceptions, such as the vaccine for whooping cough, but they’re very rare.)
Second, it’s where all the data on the Covid-19 vaccines points. “Everyone thinks the data indicate a reduction in total infections, as well as symptomatic infections,” Kilpatrick told me. “People disagree on whether we can accurately estimate how [large is] the reduction in total infections and infectiousness.”
In other words: There seems to be consensus that the vaccines don’t just keep the vaccinated safe — they make the people around them safer, too. The real question is how much safer. Lipsitch, who is more conservative than Kilpatrick at estimating that impact, still says that no effect on transmission would be “beyond shocking,”.
But the fact that the vaccines make other people safer too hasn’t necessarily made it into public messaging. News reports of the vaccine have foregrounded what the vaccine can’t guarantee and what we can’t do after we’ve been vaccinated.
“Yes, people with coronavirus vaccinations should still distance from each other. Here’s why,” argued the Washington Post.
“You’re fully vaccinated against the coronavirus — now what? Don’t expect to shed your mask and get back to normal activities right away,” begins an Associated Press story in which older people who have all been fully vaccinated are advised not to reunite with each other.
“Our discussion about vaccines has been poor, really poor,” Dr. Muge Cevik, a virologist, told the New York Times. It has overwhelmingly emphasized the fact that post-vaccine transmission is still possible, rather than frankly discussing the probability of such transmission and leaving it up to people to make their own risk calculation.
That’s because a lot of public health officials worry about encouraging people who’ve been vaccinated to “party like it’s 1999,” potentially spreading the virus to other people who haven’t had their chance to get vaccinated yet.
It’s important to note that for a vaccinated person’s behavior to be more dangerous than an unvaccinated person’s, they’d have to go really wild. If vaccines reduce infection by 90 percent, then unless your behavior gets 10 times more dangerous after you’re vaccinated, you are still safer to be around than you were before the vaccine.
Don’t go bar-hopping, but having also-vaccinated friends over is likely fine, Dr. Leana Wen of the George Washington School of Public Health argues in the Washington Post. Letting your grandparents hold your kids? Families might reasonably conclude that’s also fine, she says.
Vaccinated people should, of course, respect businesses’ rules about masks — the essential workers asked to enforce those rules have no way to know if you’ve been vaccinated. And while most people are still unvaccinated, the vaccinated should be thoughtful about protecting those who haven’t had a chance at the vaccines yet. But those reminders shouldn’t drown out an accurate understanding of the fact that the vaccines are really effective.
“Advising people that they must do nothing differently after vaccination — not even in the privacy of their homes — creates the misimpression that vaccines offer little benefit at all. Vaccines provide a true reduction of risk, not a false sense of security,” epidemiologist Julia Marcus argued in the Atlantic.
Our recommendations for vaccinated people should reflect our best current understanding of the evidence.
It’s true that there’s still some uncertainty about the magnitude of the effects of the vaccines on transmission. It’s possible that as we learn more from Israel, recommendations will change. And it’s important that people get fully vaccinated — two shots, plus some time for the immunity to fully take hold — before they assume the vaccine has fully protected them and the people around them.
But what’s important to remember is that we aren’t operating from complete ignorance. We know a lot about the vaccines, and what we know points toward them being very effective at reducing transmission and protecting those around us. If you’re hesitant about taking the vaccine because you heard that it might not protect others, you shouldn’t be, because the evidence suggests it does. That message is at least as important as warnings for the vaccinated not to “party.”
Correction, February 24: An earlier version of this article misstated the number of Americans who have received the Covid-19 vaccine. Sixty-four million doses had been administered as of the 24th, but since the vaccines approved so far are administered in two doses, 44 million people had received at least one dose by that date.