PDF version <img class="" alt="Supporters attend a campaign rally for U.S. President Donald Trump" src="//viahtml3.hypothes.is/proxy/im_///media.nature.com/lw800/magazine-assets/d41586-020-02948-4/d41586-020-02948-4_18498916.jpg"> A New Jersey campaign rally for US President Donald Trump, who has espoused herd immunity as a strategy to deal with the pandemic.Credit: Spencer Platt/Getty In May, the Brazilian city of Manaus was devastated by a large outbreak of COVID-19. Hospitals were overwhelmed and the city was digging new grave sites in the surrounding forest. But by August, something had shifted. Despite relaxing social-distancing requirements in early June, the city of 2 million people had reduced its number of excess deaths from around 120 per day to nearly zero.In September, two groups of researchers posted preprints suggesting that Manaus’s late-summer slowdown in COVID-19 cases had happened, at least in part, because a large proportion of the community’s population had already been exposed to the virus and was now immune. Immunologist Ester Sabino at the University of São Paulo, Brazil, and her colleagues tested more than 6,000 samples from blood banks in Manaus for antibodies to SARS-CoV-2.“We show that the number of people who got infected was really high — reaching 66% by the end of the first wave,” Sabino says. Her group concluded1 that this large infection rate meant that the number of people who were still vulnerable to the virus was too small to sustain new outbreaks — a phenomenon called herd immunity. Another group in Brazil reached similar conclusions2.Such reports from Manaus, together with comparable arguments about parts of Italy that were hit hard early in the pandemic, helped to embolden proposals to chase herd immunity. The plans suggested letting most of society return to normal, while taking some steps to protect those who are most at risk of severe disease. That would essentially allow the coronavirus to run its course, proponents said. Rethinking herd immunity But epidemiologists have repeatedly smacked down such ideas. “Surrendering to the virus” is not a defensible plan, says Kristian Andersen, an immunologist at the Scripps Research Institute in La Jolla, California. Such an approach would lead to a catastrophic loss of human lives without necessarily speeding up society’s return to normal, he says. “We have never successfully been able to do it before, and it will lead to unacceptable and unnecessary untold human death and suffering.”Despite widespread critique, the idea keeps popping up among politicians and policymakers in numerous countries, including Sweden, the United Kingdom and the United States. US President Donald Trump spoke positively about it in September, using the malapropism “herd mentality”. And even a few scientists have pushed the agenda. In early October, a libertarian think tank and a small group of scientists released a document called the Great Barrington Declaration. In it, they call for a return to normal life for people at lower risk of severe COVID-19, to allow SARS-CoV-2 to spread to a sufficient level to give herd immunity. People at high risk, such as elderly people, it says, could be protected through measures that are largely unspecified. The writers of the declaration received an audience in the White House, and sparked a counter memorandum from another group of scientists in The Lancet, which called the herd-immunity approach a “dangerous fallacy unsupported by scientific evidence”3.Arguments in favour of allowing the virus to run its course largely unchecked share a misunderstanding about what herd immunity is, and how best to achieve it. Here, Nature answers five questions about the controversial idea.What is herd immunity?Herd immunity happens when a virus can’t spread because it keeps encountering people who are protected against infection. Once a sufficient proportion of the population is no longer susceptible, any new outbreak peters out. “You don’t need everyone in the population to be immune — you just need enough people to be immune,” says Caroline Buckee, an epidemiologist at Harvard T.H. Chan School of Public Health in Boston, Massachusetts.Typically, herd immunity is discussed as a desirable result of wide-scale vaccination programmes. High levels of vaccination-induced immunity in the population benefits those who can’t receive or sufficiently respond to a vaccine, such as people with compromised immune systems. Many medical professionals hate the term herd immunity, and prefer to call it “herd protection”, Buckee says. That’s because the phenomenon doesn’t actually confer immunity to the virus itself — it only reduces the risk that vulnerable people will come into contact with the pathogen.But public-health experts don’t usually talk about herd immunity as a tool in the absence of vaccines. “I’m a bit puzzled that it’s now used to mean how many people need to get infected before this thing stops,” says Marcel Salathé, an epidemiologist at the Swiss Federal Institute of Technology in Lausanne.How do you achieve it?Epidemiologists can estimate the proportion of a population that needs to be immune before herd immunity kicks in. This threshold depends on the basic reproduction number, R0 — the number of cases, on average, spawned by one infected individual in an otherwise fully susceptible, well-mixed population, says Kin On Kwok, an infectious-disease epidemiologist and mathematical modeller at the Chinese University of Hong Kong. The formula for calculating the herd-immunity threshold is 1–1/R0 — meaning that the more people who become infected by each individual who has the virus, the higher the proportion of the population that needs to be immune to reach herd immunity. For instance, measles is extremely infectious, with an R0 typically between 12 and 18, which works out to a herd-immunity threshold of 92–94% of the population. For a virus that is less infectious (with a lower reproduction number), the threshold would be lower. The R0 assumes that everyone is susceptible to the virus, but that changes as the epidemic proceeds, because some people become infected and gain immunity. For that reason, a variation of R0 called the R effective (abbreviated Rt or Re) is sometimes used in these calculations, because it takes into consideration changes in susceptibility in the population. A guide to R — the pandemic’s misunderstood metric Although plugging numbers into the formula spits out a theoretical number for herd immunity, in reality, it isn’t achieved at an exact point. Instead, it’s better to think of it as a gradient, says Gypsyamber D’Souza, an epidemiologist at Johns Hopkins University in Baltimore, Maryland. And because variables can change, including R0 and the number of people susceptible to a virus, herd immunity is not a steady state.Even once herd immunity is attained across a population, it’s still possible to have large outbreaks, such as in areas where vaccination rates are low.

The R0 assumes that everyone is susceptible to the virus, but that changes as the epidemic proceeds, because some people become infected and gain immunity

“I’m a bit puzzled that it’s now used to mean how many people need to get infected before this thing stops,