The FDA’s process for evaluating the balance of risks and benefits of kids’ COVID-19 vaccines seriously underestimates benefits, and I’m worried about the message it sent to parents last week.
The October FDA Vaccines and Related Biological Products Advisory Committee meeting was contentious. Committee members include some of the world’s leading vaccinologists, pediatricians, and infectious disease experts—and they struggled mightily over extending emergency use authorization to Pfizer’s COVID-19 vaccine for kids ages 5–11.
The committee’s job is to review the evidence to date and decide whether the benefits outweigh the risks. This decision is made by vote, and the Food and Drug Administration takes it as a strong recommendation for what to do. After the FDA issues approval for the new drug to be used, it goes forward to the Center for Disease Control and Prevention’s own panel for a decision about who should get it and when. On Tuesday, the CDC panel recommended a low-dose version of the Pfizer vaccine for kids ages 5–11. Today it goes to the states, which will decide on a plan for allocating and delivering it to their residents. Some children could get vaccinated as soon as Thursday.
But I worry the FDA’s meeting has made parents doubtful. The Oct. 26 meeting was spicier than most, but it was still bureaucratic at heart, involving more than eight hours of presentations and discussion. Not to mention the 82-page briefing you get as advance reading. It’s no Game of Thrones. I never watched them live before the pandemic. It’s much more manageable simply to get the outcome after the fact.
I watched this one. I came with a lot to lose. As an academic researcher, I study herd immunity and vaccine hesitancy. And I have two kids in this age group whom I want to protect from COVID-19. I want to vaccinate them. Never mind that kids mostly don’t die when they get COVID-19; I don’t wish to have firsthand experience with the unknowns about long COVID in kids. I’m sick of the disruption of having to keep them home from school for days and stick swabs up their noses every time they get a sniffle. I want them protected. ASAP.
No doubt you’ve seen quotes from the FDA meeting flying around on social media; the sound bites were a nightmare for those of us working to stem the tide of vaccine hesitancy. The committee members used phrases like “the toughest decision,” “I’m having some challenges with this one,” and “close call.” Taken out of context, they made the vaccine sound downright dicey. For example, Dr. Eric Ruben—a member of the committee, M.D./Ph.D., Harvard professor of immunology and infectious diseases, and editor in chief of the New England Journal of Medicine—really did say, “We’re never gonna learn how safe the vaccine is until we start giving it.” This statement was, of course, taken out of context and flew far and wide on social media. It was pulled by Facebook’s fake news police bots but probably too late. Talking about the vaccine like it’s a close call in terms of being worth the risks—understandably—makes parents feel jumpy.
I couldn’t predict which way it would go based on the discussion. As I watched, my heart sank. I found myself screaming at the livestream out of frustration because no one was talking about the elephant in the room: Vaccines protect the person you give them to. And all of us.
The final discussion was especially galling. Committee member after member described the dilemma of deciding whether the benefits outweigh the risks of this vaccine. Member after member talked about how close it is, weighing clinical risks and clinical benefits in a population where both are tiny.
The FDA makes decisions about authorizing vaccines as if they’re aspirin. To decide whether Bayer should sell aspirin in the United States, we only need to know its clinical risks and benefits—the things that happen inside the body of the person who takes aspirin. Benefits: reduces pain, reduces fever, prevents heart attacks. Risks: serious gastrointestinal bleeding and other bleeding disorders. These risks and benefits must be balanced and, in this case, depend on the patient’s age and other factors. But both risks and benefits happen only to the person who took the aspirin. When we consider their balance and decide who should take a drug, we only need to consider the balance of clinical risks and benefits. There is no scenario where we expect the aspirin I take to affect my husband’s health outcomes.
But that’s just not true for vaccines. Infectious diseases such as COVID-19 are infectious. You cannot reasonably talk about the benefits of vaccines and ignore the fact that vaccination in general has a well-established track record in reducing disease transmission. Vaccines often (but not always) prevent you from getting COVID-19 in the first place, and when you don’t have COVID-19, you cannot infect anyone else. Vaccines lead to fewer and fewer transmissions over time, until there are very few. That’s how vaccines end epidemics.
We do not know precisely how much the COVID-19 vaccines reduce asymptomatic transmission (a point I’ll circle back to later), but we know they help some. In general, we do not see vaccines turning a severe infectious disease into a not-so-horrible disease that still infects just as many people. Instead, we see vaccines reducing infections overall—often to the point that the disease becomes rare in a well-vaccinated population. Vaccines can even (albeit in only one exceptional case so far) eradicate a human infectious disease from planet Earth. Reducing transmission makes vaccines the greatest weapon in the fight for public health since clean drinking water.
When the FDA looks only at clinical risks and clinical benefits, like it did during the Oct. 26 meeting, it completely ignores all the benefits at the population level. They’re pretending that children aged 5 to 11 do not live with other people. In short, they are very seriously underestimating the benefits of vaccines for this age group.
As soon as you start accounting for prevention of onward transmission, the balance immediately tips toward vaccination. For example, we have to consider benefits that accrue to the whole household. Jace Garcia of Sacramento, California, age 11, caught COVID-19 from his friend while playing soccer, then passed it along to both his fully vaccinated father and mother. It’s no surprise; any parent knows that a kid will sneeze directly into your eyeballs. There’s no way to isolate yourself from your own sick child. It’s a recipe for high-load exposures and breakthrough infections. The COVID-19 vaccines don’t always keep us from getting sick under such high-exposure conditions, but they might have kept Jace from getting sick in a casual soccer game. It’s an incremental advantage but critically important to end the pandemic.
Jace and his family are lucky; both his parents missed several weeks of work, but they recovered. What about the kid in this same situation whose mother is undergoing cancer therapy and is immunosuppressed? Or the one who lives with her grandmother? The FDA’s framing of risks and benefits ignores social, psychological, and economic risks to the child whose guardian ends up in the hospital for six weeks (and with the bills to go with it). It even ignores the unignorable loss of schooling associated with being sent home for two weeks after a COVID-19 exposure—which is happening to hundreds of thousands of children every day, nationwide.
Vaccines are not like aspirin. Vaccines are by definition intended to operate at both the clinical and the population level, protecting us from disease by stimulating our immune systems and producing immunity. Yes, they reduce the severity of a COVID-19 case even if you still end up getting COVID-19, and that’s wonderful. But they also have an effect at the population level.
This group of experts is not ignorant of population-level benefits of vaccines. (I promise, most of them have taken introductory epidemiology. Some of them have written books on vaccines.) They ignore the nonclinical benefits in this context because the FDA’s guidance instructs the committee to consider only clinical risks and clinical benefits. The voting question last week was: “Based on the totality of scientific evidence available, do the benefits of the Pfizer-BioNTech COVID-19 Vaccine when administered as a 2-dose series … outweigh its risks for use in children 5-11 years of age?”
All types of risks and benefits could be up for consideration, including clinical and population-level ones. They aren’t. The FDA provides the advisory committee members with a structured risk-benefit assessment framework, which includes a box for “risks” and another for “benefits.” The instructions for the benefits box specify members should only consider clinical benefits.
Which leads me back to one of the earlier points: Why don’t we know to what extent COVID-19 vaccination reduces asymptomatic illness and onward transmission? Answering this question during the Phase III vaccine trials would have been straightforward (if costly), but the clinical trial research team didn’t measure either asymptomatic infection or inter-household transmission. Vaccine developers didn’t include transmission in the primary trials precisely because the FDA does not consider the interpersonal benefits of vaccination important. Such data wouldn’t have helped their case for FDA approval. In fact, FDA doesn’t consider whether a vaccinated person might lead to fewer cases and ultimately the end of the pandemic at all. It evaluates vaccines using the same standards as any other drug.
The CDC’s advisory panel, called the Advisory Committee for Immunization Practices, has a different charge. It decides what population groups should get the vaccine to optimize public health, and on what dosing schedule, and its calculus does consider the broader implications of vaccinating kids. CDC Director Rochelle Walensky appeared at the beginning of the Tuesday meeting, during which her agency’s committee reviewed the Pfizer vaccine for this age group.* Walensky urged CDC committee members to consider the “detrimental social and mental health impacts” of the pandemic, “beyond the clinical impact of COVID on children.” Presentations clearly focused on both clinical and population-level benefits, and the vote was a unanimous 14-zero in favor. Discussion made it clear that when we account for the tremendous social, mental health, and other societal benefits of vaccinating kids, there was no contest. Kids ought to be vaccinated, plain and simple.
Clinical risks should always be primary in the risk-benefit analysis. Any vaccine that carries a lot of risk could never outweigh societal benefits. FDA quavered only because the clinical risks for this vaccine are extremely small. They’re so small that we struggle to put numbers on them, given the size of the clinical trials. That’s the point Ruben was making in his ill-worded statement. Given the relatively small clinical benefits in this age group, committee members weren’t sure any risk was acceptable, even the inestimably small risks of this vaccine.
Ultimately, the FDA’s committee did vote to extend Pfizer’s EUA in a 17–0 decision, with one member abstaining. I breathed a sigh of relief, but I also worried about how the committee’s discussion landed with parents. We’ll all face the decision to vaccinate our 5- to 11-year-olds very soon, and I fear the FDA’s narrow framing has helped us miss the point.
I hope that when it’s your turn to decide about the risks and the benefits of vaccinating your child, you consider this: Vaccines are not aspirin. Vaccines have extremely low risks, and yes—maybe also only small clinical benefits for kids who are otherwise low-risk.
But vaccines have more to offer than improved clinical outcomes. Vaccinating kids will protect your whole family and your neighbor who is at higher risk than you. And history tells us that it’s the way out of the pandemic. It’s not an aspirin. It’s a vaccine.
Correction, Nov. 3, 2021: Due to an editing error, this article originally misstated the day when CDC Director Rochelle Walensky met with a panel considering the authorization of the Pfizer vaccine for children. It was Tuesday, not Monday.