The mRNA COVID-19 vaccines available in the United States work. But how well they work, by what metrics and for whom … that’s all unfortunately in flux. 

Since the omicron wave crested in January, multiple studies and datasets have demonstrated that the mRNA vaccines are not nearly as effective against this variant as they were against earlier variants or the original virus. That loss of effectiveness seems to be particularly stark in children age 5 to 11.¹ While the original clinical trial data released in November reported an efficacy of 90.7 percent against infection, a report published on April 26 by the Centers for Disease Control and Prevention found that two doses of the Pfizer vaccine were only 31 percent effective at preventing omicron infection in 5- to 11-year-olds. In another study, which has not yet been peer-reviewed, the New York State Department of Health found that effectiveness against omicron infection absolutely tanked in this age group — down to just 12 percent. 

Efficacy in adults has certainly decreased, but this reduced efficacy in children is another thing entirely. As another wave of COVID laps at our feet — if it’s not already up to our knees — these numbers force us to ponder whether there’s a point where efficacy can fall so far that a vaccine is no longer worth it. And, if so, how can you even begin to decide where that point should be? 

Thinking about this stuff can feel, on the surface, like buying into an anti-vax narrative. But these are questions scientists ask themselves. And, when they do, their answers come from the biology and behavior of viruses. There is no single, universal cutoff point where a vaccine goes from vital to useless. There never has been. Instead, how good is “good enough” can vary a lot depending on how the virus works and the threats it poses to society and to specific groups of people. Right now, even with vaccinated 5- to 11-year-olds catching COVID left and right, vaccinating kids still makes sense, medical experts told me. That’s because the evidence — as well as the underlying biology — suggests that those children are still benefiting from having been vaccinated. 

When the Food and Drug Administration first solicited pharmaceutical companies to develop vaccines against SARS-CoV-2 in 2020, it set the bar at 50 percent efficacy. But according to the experts I spoke to, that 50 percent figure was more or less arbitrary. “There is no set threshold on vaccine efficacy, and the vaccines in use have a wide range of efficacies,” said Dr. William Moss, executive director of the International Vaccine Access Center at Johns Hopkins Bloomberg School of Public Health. The familiar old childhood vaccines — things like polio, measles, mumps and rubella — all have extremely high efficacies, pushing 100 percent. But we certainly use other vaccines with efficacies lower than 50 percent. In October, for instance, the World Health Organization began recommending a groundbreaking vaccine for malaria. It was about 30 percent effective at preventing severe deadly illness.

There absolutely is a hard cutoff for how much efficacy is enough to make a vaccine worthwhile, but that cutoff is going to be different from virus to virus, said Dr. Paul Offit, director of the Vaccine Education Center at the Children’s Hospital of Philadelphia. What’s a reasonable threshold to expect for one vaccine could be way too low a bar for another — or way too high. It depends on biology.

Over the past year, scientists have come to the conclusion that COVID is going to be the kind of virus whose vaccine doesn’t have a particularly high efficacy against infection. This fact — and the biology behind it — is a crucial part of understanding why scientists are still recommending that young children be vaccinated. 

A couple of things are going on here. First, mutation rate. Viruses like measles barely change at all over time. One formulation of a vaccine can work — and an individual’s immune response to it can remain effective — for many, many years. Scientists knew coronaviruses could mutate faster and more successfully than measles, but no one was really prepared for how much and how quickly SARS-CoV-2 would end up mutating, Offit said. The faster the rate of mutation and the bigger the changes, the less efficacy you can expect from a vaccine. 

The second way that virus biology affects vaccine efficacy centers around how long a virus incubates in its host before it starts to cause illness. It’s no coincidence that viruses with incubation periods measured in weeks, like measles, smallpox or rubella, have highly effective vaccines, Moss and Offit said. That’s because two types of immune responses are triggered by a vaccine. In the short term, the vaccine stimulates your body to produce virus-fighting antibodies, but those fade within three to six months. The real, long-term protection comes from memory cells, which hang out quietly until the next time you’re exposed to the virus — then they start cranking out fresh antibodies. It’s like getting a new star pitcher who’s going to burn out fairly quickly — but also getting the technology to grow clone replacements of that pitcher.

The problem, Moss and Offit said, is that the process of creating fresh antibodies takes time. If a virus incubates for a while before causing illness, then memory cells can whip up some antibodies and prevent infection. But if the incubation period is short — as it is for COVID-19 — there’s not enough time before infection sets in. The antibodies your memory cells make are still helpful in reducing the severity of the illness, though. You’d rather your clone pitcher show up late and strike out a few batters than not have a pitcher at all. 

Vaccine efficacy, then, becomes a spectrum. At one end, you’ve got rotavirus, a virus with a short incubation period — about two days — whose vaccine can’t prevent infection or spread, but it can keep babies out of hospitals, preventing serious illness at a greater than 90 percent efficacy. At the other end of the spectrum is rabies, a virus with an incubation period so long — typically two to three months — you can literally give people the vaccine after they’ve been exposed and have it prevent illness essentially 100 percent of the time. You’re just not going to get rabies-style vaccine efficacy with SARS-CoV-2, a virus with an incubation period that’s typically not much longer than that of rotavirus. 

This is why researchers like Moss and Offit think it still makes sense to give 5- to 11-year-olds a COVID vaccine, even if the efficacy against infection really is just 12 percent: The efficacy against serious illness is probably higher. 

Nobody knows exactly what the efficacy against serious illness from omicron is for kids. Serious illness from COVID is rare in kids to begin with, so the sample sizes in the two studies we have are small; the best scientists can do, then, is make estimates that come with a big error bar. The New York state study, for example, estimated that efficacy against hospitalization was 48 percent … but the actual answer could fall between -12 percent and 75 percent. 

With so few studies about this age group and such big fudge factors, Peter Muennig, a professor of health policy and management at Columbia University Mailman School of Public Health, doesn’t think we can know how effective the vaccine remains in kids at this point. But knowing that mRNA vaccines involve very low risks, Muennig said, 30 percent to 40 percent efficacy against severe illness would be enough for him to get vaccinated. Just to be safe, he said, he might even get it if the efficacy rate were 20 percent. In older kids, 12 and up, and in adults, effectiveness against serious illness remains well above that. A recent analysis by the Kaiser Family Foundation found that being vaccinated had reduced adults’ chances of dying from COVID — even after omicron had become dominant – by 10 times compared with unvaccinated adults. Those who had gotten a booster reduced their risk of death by 20 times. “If it reduces chances of death [that much], that’s a no-brainer to me,” Muennig said. There’s good reason to suspect that the real efficacy against severe illness in kids is lower than that, but still high enough that getting vaccinated makes sense. It certainly would if that estimate of 48 percent efficacy is right.

Ultimately, these scientists told me, the decision comes down to weighing the risks and benefits not just of a vaccine but of that vaccine to a specific segment of the population. And that can legitimately get tricky when you’re talking about a group, such as kids, that has very low risks from the virus to begin with. But, Moss said, with variants like omicron spreading so rampantly through so much of the population, even small dangers from infection can add up. Since omicron took hold, children have accounted for only about 3 percent of COVID hospitalizations — but that’s still more than 12,500 hospitalized kids between the week of Dec. 30 and the week of April 28. And the cumulative population-based hospitalization rate in unvaccinated kids was double that of vaccinated kids during the omicron wave. 

Moss doesn’t think there’s been enough discussion on whether you need a higher efficacy rate for a vaccine in kids than in adults if the virus is less dangerous to kids. But we know the dangers of COVID-19 are big enough — and the vaccine efficacy estimates are high enough — that he has no problem recommending that kids get the jab.