Unforeseen safety issues routinely emerge after any new medicine or vaccine goes from testing in tens of thousands of volunteers to actual public use on tens of millions. So it was no major surprise when an extremely small percentage of people developed a strange blood clotting problem after receiving either the Johnson & Johnson (J&J) COVID-19 vaccine or the AstraZeneca shot, which is widely used outside of the U.S.

Rare but dangerous side effects from vaccines can present a tricky dilemma for public health authorities. In this case, the life-threatening blood clots, accompanied by an oddly low count of clot-promoting platelets, appear to strike about two individuals per million people vaccinated with J&J’s shot and about one per 100,000 receiving AstraZeneca’s. Both are minuscule risks, compared with COVID-19 itself, which, by one estimate, kills roughly two people out of 1,000 infected (though fatality rates vary greatly by age, location and other factors). On one hand, it is crucial to be transparent with the public—and to alert health care providers to the problem and advise them on how best to identify and treat it. On the other hand, there is a chance of sowing unwarranted doubts about these vaccines and perhaps others as well, inflaming already worrisome levels of vaccine hesitancy.

“The minute you’ve told people that there’s a risk, even if it’s one in a million, I think what they hear is ‘That could happen to me,’” says pediatrician and vaccine researcher Paul Offit, director of the Vaccine Education Center at Children’s Hospital of Philadelphia.

The discovery of the clotting issues—in early March for AstraZeneca and early April for J&J—prompted emergency meetings of health officials in the U.S. and Europe and pauses in distribution of both vaccines in countries around the world. But by April 23 public health authorities in both regions concluded that the benefits of these vaccines far exceed the risks and that distribution should resume, albeit with new labels warning about the very rare blood clots.

Whether the pauses were justified is a matter of debate, given the urgent need to stem a deadly global pandemic. But as a question of pure mathematics, experts say the answer is clearly no. “Out of a theoretical million people who get this vaccine,” just a few will suffer the peculiar clots, Offit points out. “But out of a theoretical million people who get COVID, thousands will die.”

One of the toughest questions for health authorities is a psychological one: To what degree can the public make practical sense of a very rare—but very frightening—possible side effect? “Most people are risk-illiterate,” says psychologist Gerd Gigerenzer, director of the Harding Center for Risk Literacy at the University of Potsdam in Germany. “A large number of studies show that ordinary people are confused by relative risks, odds or proportions.” Indeed, research demonstrates that we are particularly apt to overestimate the probability of a rare event if it is new and deadly and has been amplified by the news media, says psychologist Baruch Fischhoff, a professor of engineering and public policy at Carnegie Mellon University and an authority on how to communicate health risks.

Both Fischhoff and Gigerenzer believe health officials can help avoid such confusion by providing very clear, well-formulated information to the public—something Fischhoff says U.S. and European health authorities could be doing much better. “Most people have no problem understanding risk if you, the expert, do your job right,” he insists. “Give math a chance!”

Math Challenge

Word of the clotting problem first began filtering out in late February in connection with the AstraZeneca vaccine, which is not yet approved in the U.S. By March 22 the European Medicines Agency (EMA) was aware of 86 cases, 18 of them fatal, in Europe and the U.K.—an extremely small number relative to the 25 million people who had received the vaccine at that point. The cases were concentrated in women below the age of 60.

This month U.S. health authorities began receiving reports of similar cases linked to the J&J injection, a single-dose vaccine that has been given to about eight million people in the country. At least 15 clotting cases have been reported, all in women younger than 60, and at least three have been fatal.

In Europe, the AstraZeneca findings triggered a patchwork response. More than 20 countries stopped distributing that vaccine for a week or more. Most resumed, with an array of new recommendations that the vaccine be used only for adults older than 55, 60 or 65—or, in the case of the U.K., that adults under age 30 should opt for other vaccines. The EMA, however, has not endorsed any age restrictions, nor has it stated that younger women should avoid choosing AstraZeneca—positions mirrored by U.S. authorities with respect to the J&J vaccine.

In the U.S., the J&J pause lasted 10 days. Leaders at the Centers for Disease Control and Prevention and the Food and Drug Administration explained that they needed time to investigate the blood clot cases and were acting “out of an abundance of caution.”

That phrase did not sit well with a number of health communication experts. “It makes it sound like there’s no downside [to the pause], when in fact there can be a significant downside,” says Glen Nowak, director of the Center for Health and Risk Communication at the University of Georgia and former chief of vaccine communication at the CDC. Among the biggest concerns was the impact of losing the one-dose, easy-to-transport J&J vaccine for use in rural communities, people experiencing homelessness and other hard-to-reach groups. “We were using the J&J vaccine with vulnerable populations and transient populations, where people might not be able to come back for a second dose,” Nowak says.

For the public to make sense of such decisions, Fischhoff says, health officials have a duty to explain, with specific numbers, whether fewer people will likely be harmed with a pause than without it—an estimate he says they can reasonably make, even though it will vary from place to place. They should also explain how confident they are in their information. In his view, initial explanations for the vaccine pauses were “a colossal communications failure.” As of publication time, the CDC has not responded to Scientific American’s request for comment.

Both the CDC and the EMA did release some information about relative harm. At an April 23 virtual media event explaining the decision to resume using the J&J vaccine, CDC director Rochelle P. Walensky noted that every one million doses of J&J vaccine given to women ages 18 to 49 would prevent 650 hospitalizations and 12 deaths from COVID while possibly causing about seven cases of blood clots. The EMA has gone even further: it published a detailed (if somewhat daunting) set of infographics comparing such statistics for the AstraZeneca vaccine, breaking them down by age and by the prevalence of the coronavirus. For instance, in places with a high incidence of infection, a million doses of the vaccine given to adults in their 30s would prevent 81 hospitalizations for COVID and could involve 1.8 cases of clotting. For people in their 80s, the estimates were 1,239 prevented hospitalizations and 0.4 clotting cases.

Can most people follow this kind of math? Gigerenzer’s work has shown that presenting information in simply written tabular form—what he and his associates call “fact boxes”—can help people more easily weigh risks against benefits for vaccines and other health interventions. Well-crafted infographics that visually demonstrate relative risk, such as those created by the Winton Center for Risk and Evidence Communication at the University of Cambridge, which were the models for the EMA’s charts, also help. In the longer term, Gigerenzer believes much more can be done to educate the public, beginning in childhood, in how to think about risk and probability. And he notes that there is another lesson that would go a long way toward helping people grasp what a one-in-a-million side effect signifies: we need to be taught that almost nothing comes without risk.

At the April 23 news conference, the CDC’s Walensky acknowledged the urgent need for good communication to help people assess vaccine benefits and risks. “We have to do extraordinary outreach to patients, to meet people where they’re at, to educate them,” she said.

The ultimate impact of the pauses and of the uneven messaging about the side effects remains to be seen. But there is reason to worry, says Heidi J. Larson, director of the Vaccine Confidence Project, a research organization that tracks global views of vaccines. In the weeks after European nations suspended use of the AstraZeneca vaccine, she says, “we saw a precipitous drop in vaccine confidence in Africa”—where that shot had been expected to become a mainstay. A Washington Post–ABC News poll, conducted during the J&J pause, suggests similar damage has been done in the U.S.: only 22 percent of unvaccinated Americans said they would be willing to take the J&J shot. Larson and others worry that confused and concerned people often turn to unreliable sources on vaccine safety. “If they’re not getting clear answers,” Larson says, “it just leaves the gate wide open for misinformation and disinformation. People fill that space with their own story line.”

Read more about the coronavirus outbreak from Scientific American here. And read coverage from our international network of magazines here.