Several months into a pandemic that has claimed hundreds of thousands of lives and decimated economies around the world, scientists still lack a complete understanding of how the virus that caused it is transmitted. Lockdowns are already easing in some places, and people are preparing to return to a version of work and social life. But a crucial question stubbornly remains: Can the pathogen behind COVID-19 be “airborne”?

For months, the U.S. Centers for Disease Control and Prevention and the World Health Organization have maintained that the novel coronavirus is primarily spread by droplets from someone who is coughing, sneezing or even talking within a few feet away. But anecdotal reports hint that it could be transmissible through particles suspended in the air (so-called "aerosol transmission"). And the WHO recently reversed its guidance to say that such transmission, particularly in “indoor locations where there are crowded and inadequately ventilated spaces where infected persons spend long periods of time with others, cannot be ruled out.” After attending a choir practice in Washington State in early March, dozens of people were diagnosed with or developed symptoms of COVID-19 even though they had not shaken hands or stood close to one another. At least two died. After dining at an air-conditioned restaurant in China in late January, three families at neighboring tables became sickened with the virus—possibly through droplets blown through the air.

To address the prospect of airborne spread of the novel coronavirus, it is first necessary to understand what scientists mean by “airborne.” The term refers to transmission of a pathogen via aerosols—tiny respiratory droplets that can remain suspended in the air (known as droplet nuclei)—as opposed to larger droplets that fall to the ground within a few feet. In reality, though, the distinction between droplets and aerosols is not a clear one. “The separation between what is referred to as ‘airborne spread’ and ‘droplet spread’ is really a spectrum,” especially when talking about relatively small distances, says Joshua Santarpia, an associate professor of pathology and microbiology at the University of Nebraska Medical Center.

Airborne spread has been hypothesized for other deadly coronaviruses, including the ones that cause severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS). A handful of studies suggest the new coronavirus, SARS-CoV-2, can exist as an aerosol in health care settings. But much remains unknown about whether the aerosolized virus is infectious and what amount of virus one needs to be exposed to in order to become sick, known as the minimal infectious dose. Even if aerosol transmission does occur, it is not clear how common it is, compared with other transmission routes, such as droplets or surfaces. Uncovering this information is vital, especially given the fact that people can spread SARS-CoV-2 when they have no symptoms.

Perhaps “Is the coronavirus airborne?” is the wrong question. COVID-19 may have the potential for airborne spread, says Stanley Perlman, a professor of microbiology at the University of Iowa. “But whether [this route is] important clinically is really the question one wants to know about,” he says.

Evidence for Aerosol Transmission

Some of the strongest evidence that airborne transmission of the new coronavirus may be possible comes from a study published late last month in Nature. In it, researchers measured the virus's genetic material, or RNA, in aerosols sampled in February and March at two hospitals in Wuhan, China—the city where the outbreak is widely believed to have begun. The researchers found very low levels of airborne viral RNA in the hospitals’ isolation wards and in ventilated patient rooms. But there were measurably higher levels in some of the patients’ toilet areas. They also found high levels of viral RNA in places where medical workers remove protective gear, as well as in two crowding-prone locations near the hospitals. “Our study and several other studies proved the existence of SARS-CoV-2 aerosols and implied that SARS-CoV-2 aerosol transmission might be a nonnegligible route from infected carriers to someone nearby,” says study co-author Ke Lan, a professor and director of the State Key Laboratory of Virology at Wuhan University.

A preprint (not yet published) study led by Santarpia and his colleagues similarly found evidence of viral contamination in air samples and surfaces from rooms where COVID-19 patients were being kept in isolation. “I think there are a lot of us—myself included—who feel very strongly that the airborne route of transmission is very possible,” he says. “I would hesitate to call it proven by any means. But I think there’s mounting evidence to support it.”

Both the Nature study and Santarpia’s paper measured viral RNA, not actual virus, so it is not clear that the material found in aerosols was functionally infectious. “Finding RNA doesn’t tell you [that] you have aerosol spread,” says Perlman, who was not involved in either study.

Another paper, recently published in the New England Journal of Medicine, showed that infectious SARS-CoV-2 virus can remain in aerosols for at least three hours—and for several days on various surfaces—in a laboratory setting. But the amount of viable virus diminished significantly during that time. Scientists do not know the infectious dose of SARS-CoV-2. (For influenza, studies have shown that just three virus particles are enough to make someone sick.)

Overall, most of the evidence that SARS-CoV-2 can become airborne comes from clinical settings—which tend to have a lot of sick people and may host invasive procedures, such as intubations, that can cause patients to cough, generating aerosols. It is not clear how representative of everyday environments these areas are. “There is not much convincing evidence that aerosol spread is a major part of transmission” of COVID-19, Perlman says.

That assessment does not mean it is not occurring, however. Benjamin Cowling, head of the division of epidemiology and biostatistics at the University of Hong Kong’s School of Public Health, says there is a popular misconception that if a virus can spread through the air at all, it must be able to spread over a long range. He gives the analogy of being in a restaurant where someone is smoking: “If the person on the other side of restaurant is smoking, you probably wouldn’t smell it, and you’d never even notice. That’s because the smoke would never reach you,” he says. “It doesn’t mean there’s not smoke produced.” In other words, just because SARS-CoV-2 may not be transmitted over a long range, that does not mean it is not airborne. Like cigarette smoke, aerosol particles spread around a person in a cloud, with the concentration being highest near the smoker and lower as one gets farther away.

Even if aerosols do not travel farther than most droplets, the oft-touted “six-foot rule” for social distancing may depend on the circumstances, Cowling says. If there is a fan or air conditioner, infectious aerosols (or even droplets, as was suspected in the case of that restaurant in China) could potentially sicken someone farther away who is downwind.

Some evidence suggests that talking could be a significant mode of viral transmission. A study published on May 13 in Proceedings of the National Academy of Sciences USA used laser light scattering to visualize tiny saliva droplets expelled during speech. The research did not measure droplets with viable SARS-CoV-2 virus. But if one assumes the droplets contain seven million virus particles per milliliter, a minute of loud speech could generate more than 1,000 virus-containing droplets that could hang in the air for eight minutes or more, the researchers write in the study. “There is a substantial probability that normal speaking causes airborne virus transmission in confined environments,” they conclude.

Factors That Affect the Risk of Airborne Spread

Cowling hypothesizes that many respiratory viruses can be spread through the airborne route—but that the degree of contagiousness is low. For seasonal flu, the basic reproduction number, or R0—a technical designation for the average number of a people a sick person infects—is about 1.3. For COVID-19, it is estimated to be somewhere between two and three (though possibly as high as 5.7). Compared with measles, which has an R0 in the range of 12 to 18, these values suggest most people with the disease caused by SARS-CoV-2 are not extremely contagious.

But there are seeming exceptions, such as the choir practice in Washington State, Cowling says. A CDC report about the event released on May 12 found that of the 61 people who attended the 2.5-hour practice (one of whom had coronavirus symptoms), 32 developed confirmed COVID-19 infections and 20 developed probable ones. The report concluded that “transmission was likely facilitated by close proximity (within 6 feet) during practice and augmented by the act of singing” and that singing “might have contributed to transmission through emission of aerosols, which is affected by loudness of vocalization.” For unknown reasons, some individuals seem to infect many more people than others do. These so-called superspreaders were documented in the SARS outbreak of 2003, too. In what has become known as the 20/80 rule, about 80 percent of infectious-disease-transmission events may be associated with just 20 percent of cases, Cowling notes. “We don’t know how to identify those 20 percent,” he says. “But if we were able to, in some way, then that would be a major advance.”

Ventilation likely also plays an important role in how easily the virus can be transmitted through the air. Indoor spaces probably pose a higher risk than outdoor ones, especially if they are poorly ventilated, Cowling and others say. Crowded areas such as bars, restaurants and subway trains could all be risky—especially if people are asymptomatic and spend long periods of time in such areas. Precautions could include better ventilation, regular cleaning and mask wearing.

Cowling co-authored a study, published in early April in Nature Medicine, of patients with respiratory infections at an outpatient clinic in Hong Kong between 2013 and 2016. This research detected RNA from seasonal coronaviruses—the kind that cause colds, not COVID-19—as well as seasonal influenza viruses and rhinoviruses, in both droplets and aerosols in the patients’ exhaled breath. The paper, led by Nancy Leung, an assistant professor at the University of Hong Kong’s school of public health, found that wearing surgical masks reduced the amounts of influenza RNA in droplets and of seasonal coronavirus RNA in aerosols.

Although the study did not look at COVID-19 specifically, the findings support mask wearing as an effective way to limit transmission of the virus from an infected person—known in medical parlance as source control. There is not much evidence that masks convey protection to healthy people, although it is possible (and may depend on the type of mask). Given the prevalence of asymptomatic infection with COVID-19, however, there is some justification for universal mask wearing to prevent those who do not know they are sick from infecting others. In Hong Kong, which has kept its outbreak relatively under control, masks are worn by the vast majority of the population, Cowling says.

The likelihood of airborne transmission—especially compared with other routes, such as droplets or surfaces—remains unclear. Most researchers still think the new coronavirus is primarily spread via droplets and touching infected people or surfaces. So diligent hand washing and social distancing are still the most important measures people can take to avoid infection.

Leung puts the risks in perspective. Most of what people know about aerosol transmission is from tuberculosis, measles and chickenpox, she says—and these pathogens usually have high transmissibility, with the potential for long-range spread. “The conventional thinking is, therefore, once you mention there’s aerosol transmission, everyone is so worried because [they assume that the virus has] higher transmissibility and that it’s more difficult to control,” she explains. But even if there is airborne transmission, it may only happen at short range—within which other infection routes may be just as likely—or more so. Thus, Leung adds, “having a higher risk of aerosol transmission itself doesn’t necessarily translate to more transmissibility.”

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

Editor's Note (7/9/20): This article was updated in light of the World Health Organization's acknowledgment of potential aerosol transmission of the new coronavirus. The reversal came after more than 200 scientists contributed to an open letter pointing to the cumulative evidence for this type of transmission. The text had previously been amended on May 15 to include new information about droplets spread by talking and about a choir practice in Washington State that sickened many people.