Public confidence in the safety of COVID-19 vaccines dipped in the United States after government officials paused vaccinations with the Johnson & Johnson (J&J) shot last month, according to a poll . During the ten-day hiatus, officials explored whether the vaccine was linked to a rare type of blood clot , but they ultimately deemed the jab safe and gave the green light to resume its use. After the pause began, 7% of unvaccinated adults who were surveyed said that the news about blood clots made them less likely to want any COVID-19 shot, according to data published by the Kaiser Family Foundation (KFF).

The incident demonstrates the knife-edge that public-health authorities have been walking since COVID-19 vaccines became available late last year. Vaccines represent a way to end the COVID-19 pandemic and to protect individuals, so authorities would like people to sign up for them. But the shots come with risks that need to be communicated transparently, to maintain trust and uphold scientific ethics, says Hilda Bastian, an independent scientist who studies evidence-based medicine in Victoria, Australia.

Public-health specialists must always strike a careful balance when communicating about vaccine safety, but the enormous scale of the COVID-19 vaccine roll-out means that safety data are evolving fast — so researchers are scrambling to share developments transparently and clearly with the public. And they worry that with the rise of anti-vaccination movements, their messages might be used or interpreted to fuel misinformation campaigns. Those who spoke to Nature say that because the stakes are so high for COVID-19, explaining vaccine risk has been especially fraught. “It’s a minefield,” says Bastian.

Kathryn Edwards, a vaccinologist at Vanderbilt University School of Medicine in Nashville, Tennessee, who has been a vaccine-safety consultant for 40 years, agrees. “I have not worked harder in my life,” she says.

The effects of a pause

US officials paused inoculations with the J&J vaccine last month so that researchers at the US Centers for Disease Control and Prevention and the US Food and Drug Administration could evaluate 6 cases of a rare type of blood clot reported among 6.8 million people who had received a J&J jab. The pause followed an announcement by the European Medicines Agency that linked the Oxford–AstraZeneca COVID-19 vaccine to reports of a similar blood-clotting condition.

By the end of the pause, US officials had identified 15 cases of a rare but severe condition called thrombosis with thrombocytopaenia syndrome (TTS) in people who had received the J&J vaccine. The highly specific hallmarks of TTS are clots in unusual parts of the body, such as the brain or abdomen, combined with low blood-platelet levels. The cases occurred exclusively in women between the ages of 18 and 59. Still, officials decided that the benefit of protection against COVID-19 imparted by the vaccine outweighs the remote risk of TTS, so they lifted the pause and asked health-care providers to update information provided with the J&J shot to include warnings about the condition.

During the pause and the week after it was lifted, the KFF, a non-profit health-policy organization based in San Francisco, California, polled around 2,100 adults across the United States about whether it had affected their stance on COVID-19 vaccines. Although 69% of respondents expressed confidence in the other two vaccines being administered in the country — the Pfizer–BioNTech and Moderna shots — only 46% said the same for J&J. Among the subset of respondents who had not yet been vaccinated, roughly 20% said that the pause had changed their view of at least one of the COVID-19 vaccines in some way (see ‘New hesitancy’).

Given the intense news coverage of safety concerns around J&J, “it’s not surprising that people might have questions”, says Katherine Schaff, who studies public-health communication at Berkeley Media Studies Group in California. She emphasizes that confidence in the other available COVID-19 vaccines remains high, and the J&J pause is ultimately evidence of vaccine safety-monitoring systems working as they should. But Schaff adds that the new hesitancy described in the KFF poll is evidence that more can be done to make sure people have access to transparent communication about their safety.

Risk is not only a number

One challenge public-health authorities face is putting risk into context without seeming to dismiss people’s worries, says Heidi Larson, an anthropologist at the London School of Hygiene & Tropical Medicine who specializes in risk and decision science. Even if authorities say the likelihood of a severe reaction is one in a million, she says, what people want an answer to is, “What does that one in a million mean for me or someone in my family?”

Providing that context can be tricky because risk perception is highly subjective, says Alexandra Freeman, executive director of the Winton Centre for Risk and Evidence Communication at the University of Cambridge, UK. There are two elements of risk that people need to understand to make decisions, she says: the likelihood of something happening, and the impact of something happening. For example, the likelihood of severe, influenza-like symptoms after a vaccine injection might be one in ten, but if they happen, those symptoms could have a larger impact for a single parent without childcare support than for someone able to take time to recover.

Public-health specialists told Nature that the key to increasing public trust continues to be transparency. In a study of how communication about vaccine efficacy affected people’s decisions to get a COVID-19 shot ¹ , Freeman and her colleagues found that being transparent about the uncertainties made no difference to whether or not a person got a vaccine. But, says Freeman, “we did find that people felt that they were more informed and felt more confident in their decision making when they were given more informative communication”.