“No definitive clinical trial can give you this evidence,” Rasmussen said. The trials were really designed for speed and safety, so the researchers were most concerned with looking for symptomatic COVID-19 or adverse reactions, not asymptomatic infections. To know how often vaccinated people were asymptomatically carrying the virus, researchers would have had to test each of the tens of thousands of people in their clinical trials as frequently as possible.
Some ongoing trials have taken to swabbing the noses of vaccinated people occasionally, and this could add insight into how common it is for people to carry the virus after vaccination. Early evidence from Johnson & Johnson’s clinical trial, for example, suggests a significant reduction in transmission after vaccination, though this remains to be verified. Still, occasional testing is bound to miss cases of infection, and finding some virus in some noses doesn’t tell us how infectious the owners of those noses might be—or whether they’re infectious at all.
The only way to answer this question for certain would be to run a “challenge” trial in which vaccinated and unvaccinated people were deliberately exposed to the virus under similar conditions, and then tested to see what percentage of them got infected. That’s just step one. Then the vaccinated-but-infected people would need to hang out with a bunch of unvaccinated people to see if they got infected, and at what rate. This is not going to happen. Challenge trials are ethical minefields in normal times; at this point, any study that involves withholding a vaccine from a control group would be difficult to justify.
More trial data are expected over the next few months, and these may help narrow our uncertainty. It would certainly be useful to get a better sense of whether the risk of catching COVID-19 from your grandmother, for example, drops by something like 90 percent once she’s vaccinated, or whether it’s closer to 10 percent—but that number isn’t going to be exact, and it won’t be static, either. Even if we could somehow run the sort of challenge trial described above, whatever value it produced could change as new variants of the virus take hold, and it might well vary across regions with different patterns of prior infection, behavioral norms, local weather, and other variables we don’t even know to look for.
All of this is academic. Whatever trial data might arrive in the coming months won’t change the practical advice: As long as a lot of virus is still circulating in a community and many people remain unvaccinated, the mere fact that some have protection will not mean that it’s responsible for them to forgo precautions and do whatever they like.
A different kind of data, though, will offer that reassurance and certainty. This is what we’re really waiting on. “We will absolutely get to a point when we can say that vaccinated people don’t need to wear masks,” Madad said, but that will be driven largely by changes in the number of cases, and in the vaccination rate. The sooner we can drive the former down and the latter up, the sooner normalcy returns. As populations draw closer to herd immunity, the chance of a vaccinated person both carrying the virus and coming into close contact with a nonimmune person will become so low that the guidelines will change. But as long as the virus remains omnipresent, the risk of getting infected (and transmitting) the virus after being vaccinated remains too high to countenance.