New studies hint that the coronavirus may be evolving to become more airborne

The virus appears to spread through the air, but masks reduce the amount of infectious virus

a researcher sits crosslegged on a hospital bed with their face inside a silver cone that's connected to a transparent chamber full of scientific apparatus

A researcher demonstrates the use of a device that captures fine aerosols and large droplets in a person’s breath exhaled while breathing, talking and singing.

National University of Singapore

Small aerosol particles spewed while people breathe, talk and sing may contain more coronavirus than larger moisture droplets do. And the coronavirus may be evolving to spread more easily through the air, a new study suggests. But there is also good news: Masks can help.

About 85 percent of coronavirus RNA detected in COVID-19 patients’ breath was found in fine aerosol particles less than five micrometers in size, researchers in Singapore report August 6 in Clinical Infectious Diseases. The finding is the latest evidence to suggest that COVID-19 is spread mainly through the air in fine droplets that may stay suspended for hours rather than in larger droplets that quickly fall to the ground and contaminate surfaces.

Similar to that result, Donald Milton at the University of Maryland in College Park and colleagues found that people who carried the alpha variant had 18 times as much viral RNA in aerosols than people infected with less-contagious versions of the virus. That study, posted August 13 at, has not been yet been peer-reviewed. It also found that loose-fitting masks could cut the amount of virus-carrying aerosols by nearly half.

In one experiment, the Maryland team grew the virus from the air samples in the lab. That could be evidence that may convince some reluctant experts to embrace the idea that the virus spreads mainly through the air.

The debate over aerosol transmission has been ongoing since nearly the beginning of the COVID-19 pandemic. Last year, 200 scientists wrote a letter to the World Health Organization asking for the organization to acknowledge aerosol spread of the virus (SN: 7/7/20). In April, the WHO upgraded its information on transmission to include aerosols (SN: 5/18/21). The U.S. Centers for Disease Control and Prevention had acknowledged aerosols as the most likely source of spread just a few weeks before.

Previous studies in monkeys have also suggested that more virus ends up in aerosols than in large droplets. But some experts say that direct evidence that the virus spreads mainly through the air is still lacking.

“There’s lots of indirect evidence that the airborne route — breathing it in — is dominant,” says Linsey Marr, a civil and environmental engineer at Virginia Tech in Blacksburg, who studies viruses in the air. She was one of the 200 scientists who wrote to the WHO last year. “‘Airborne’ is a loaded word in infection control circles,” she says, requiring health care workers to isolate patients in special rooms, wear protective equipment and take other costly and resource-intensive measures to stop the spread of the disease. For those reasons, infection control experts have been reluctant to call the coronavirus airborne without especially strong proof.

Most COVID-19 cases have been among close household contacts — typically within the 6-feet splash zone of large droplets. It can be hard to tease out whether such infections were passed on by large droplet contamination or by breathing the same air. But for other situations, such as when patrons get infected while sitting across a restaurant from someone with COVID-19, aerosols are really the only explanation, Marr says.

Mechanical engineer Kwok Wai Tham of the National University of Singapore set out to sample how much virus COVID-19 patients produce when they breathe, talk or sing, in part, to address skeptics’ concerns. “I’m doing this to convince some very close friends,” he says. He and colleagues rolled a mobile lab into 22 patients’ rooms and had volunteers stick their heads into a large metal cone.

The researchers collected both aerosols and larger droplets that the patients exhaled while breathing quietly for 30 minutes, while repeating passages from Dr. Seuss’ Green Eggs and Ham for 15 minutes, or while singing simple tunes like the “Happy Birthday” song, “Twinkle, Twinkle Little Star” or the “ABCs” for 15 minutes. The scientists tested both aerosols and large droplets in the air samples for coronavirus RNA and calculated how many copies of the virus’s nucleocapsid protein gene, or N gene, were present. That gives an estimate of how much virus is in a sample.

Of the 22 patients who sang for science, only 13 spewed forth detectable levels of viral RNA. In general, singing created the most virus-laden aerosols, but some people generated more while talking. Those differences might be attributable to the volume at which volunteers sang, Tham says. “Some people were shy and sang softer. Others were quite uninhibited.”  

The overall amount of virus that people produced varied widely. Scientists already knew that some people are more likely to spread the virus than others, including some people involved in superspreading events (SN: 6/18/20). In this new study, the differences weren’t due to symptoms — some asymptomatic people made more virus than those with fevers, coughs or runny noses.

Only one factor stood out as affecting the amount of virus emitted. People who were earlier in the course of infection tended to produce more virus, the researchers found. That agrees with data from lab animal studies and other human studies suggesting that people are most contagious in the first week after catching the coronavirus (SN: 3/13/20).

So far, Tham’s skeptical virologist friends aren’t convinced that he’s demonstrated that aerosol transmission is the major route of COVID-19 spread. “They say, ‘we need the golden evidence. Show me a live virus that is retrieved from the air,’” Tham says.

Viral RNA could be debris from dead viruses that can’t cause infection, says Andrew Pekosz, a virologist at the Johns Hopkins Bloomberg School of Public Health who was not involved in either study. “In the absence of infectious virus, the significance of aerosols on transmission is still a bit unclear.”

The study from the Maryland group may provide that evidence. In that study, people with asymptomatic or mild coronavirus cases recited the ABCs, shouted “Go Terps!” (the Maryland mascot) or sang “Happy Birthday” into a similar device. In this study, the infected people did the activities once while wearing a mask and once while not wearing one.

About 45 percent of fine aerosol particles contained viral RNA, as did 31 percent of coarse aerosols larger than 5 micrometers and 65 percent of swabs collected from the volunteers’ mobile phones, the researchers found. (In medical research, objects such as phones that can carry disease-causing organisms are known as fomites.)

In addition, the increased amount of alpha variant in aerosols may suggest that the coronavirus is evolving toward more efficient airborne spread, the researchers propose. The study was done from May 2020 to April 2021, before the delta variant began its surge in the United States.

Researchers were able to grow infectious virus from two of 66 aerosol samples, both collected while people were wearing masks. None of the coarse aerosols or phone swabs yielded any infectious virus.

Although the Maryland group used an efficient way to look for infectious virus in aerosols, it was still rare to find them, Pekosz says. “It would be difficult to make the case that this was what is responsible for increased spread of alpha.”

But Marr says the data do suggest the coronavirus is evolving toward more efficient spread through the air. Although the study involved only four patients infected with alpha, those people consistently released more virus than people infected with other variants. “These results combined with epidemiological observations about the spread of alpha, and now delta, support the idea that these variants are supercharged when it comes to aerosol transmission,” she says.

The masks volunteers wore in the Maryland study were mostly loose-fitting. They ranged from a single-layer homemade cloth mask early on and progressed over the course of the study to double-layer commercially made cloth masks, to double masks, surgical masks and one KN95 mask by the end. On average, the masks reduced the number of virus-containing, coarse aerosols produced by 77 percent compared with no mask. And virus-laden fine aerosols were reduced an average of 48 percent, though the reduction ranged from 3 percent to 72 percent. Masks performed equally well against the alpha variant as for other variants.  Previous studies have suggested that well-fitting masks — ones that seal tightly to the face and don’t leave gaps at the tops, bottoms or sides for the virus to pass unfiltered — may reduce coronavirus exposure by 96 percent if everyone is wearing them (SN: 2/12/21).

The latest results suggest that masks can help reduce the amount of virus people give off, though the coronavirus can still escape if the face coverings are worn loosely. “With the dominance of newer, more contagious variants than those we studied, increased attention to improved ventilation, filtration, air sanitation, and use of high-quality tight-fitting face masks or respirators … will be increasingly important for controlling the pandemic,” the researchers wrote. That’s especially important in places with low vaccination rates.

More Stories from Science News on Health & Medicine

From the Nature Index

Paid Content