The study appears to be more bad news for restaurants, which have already been identified in research as a primary source for the spread of the virus. The Korean researchers recommend that public health authorities update safety guidelines based on their study, arguing that six feet of space between tables is not enough to protect diners from being infected.
At present, the Centers for Disease Control and Prevention recommends that diners wear a mask when “less than 6 feet apart from other people or indoors.” The agency also recommends diners wear a “mask as much as possible when not eating” and maintain “a social distance of 6 feet or more in any entryway, hallway or waiting area.” For the first time, the CDC recently recommended “universal mask use” outside people’s homes.
But as infection rates and hospitalizations spike across the country, local and state jurisdictions have already begun to clamp down on restaurants, including New York City, which will prohibit indoor dining starting Monday in a metropolis where nearly nine out of 10 eateries already could not pay their full rent this summer. The South Korean study would seem to give public health officials more ammunition to shut down indoor dining until the pandemic passes.
Droplet “transmission can occur at a distance greater than 2 [meters] if there is direct air flow from an infected person in an indoor setting,” the authors write. “Therefore, updated guidelines for quarantine and environmental management of covid-19 are needed until approval of an effective treatment drug or vaccine.”
While some have praised the scientific rigor of the research, others have cautioned that you can draw only limited conclusions from a report that’s more of a small, three-person natural experiment than a larger scientific inquiry on the conditions of air flow inside restaurants and how they can spread the virus.
“The problem that you tend to have is one of missing information,” says Richard Martinello, an associate professor at Yale School of Medicine and a specialist in adult and pediatric infectious diseases.
“They may know well what happened within that restaurant,” Martinello adds, “but they don’t know what happened on the sidewalk outside the restaurant. They don’t know what happened back in the kitchen at the restaurant. There are so many other aspects,” including the fact that one in five people infected with the coronavirus will experience no symptoms, but may still spread the virus.
Yet the South Korean researchers did have mountains of data at their disposal, more than their counterparts in the United States could probably ever get their hands on. South Korea was quick to use technology — including cellphone location data, credit card records and closed-circuit television feeds — to try to prevent the spread of the virus. As a result, the South Korean government has built an extraordinary database on coronavirus cases, the kind that would raise legal, privacy and ethical concerns in America.
“We probably could not do this study in the United States,” says Donald Milton, professor of environmental health at the University of Maryland School of Public Health. Milton was among the experts who argued early in the pandemic that the coronavirus was spread through smaller aerosol particles, not just larger person-to-person droplets.
Using the available data, including the sequencing of the viral genome isolated from infected diners, the Korean authors were able to pinpoint the location where they think diner A (a high school student who had not traveled outside her hometown of Jeonju, South Korea) was originally infected on June 12. It was at a first-floor restaurant in Jeonju, where there had not been a coronavirus infection in the two months preceding this case, according to a recent Los Angeles Times story. The restaurant has no windows or ventilation system, but has two ceiling air conditioners that circulated air in the direction of the two diners (A and C) who were then infected from diner B (a door-to-door saleswoman visiting from another town, according to the L.A. Times).
The high school student and her companion entered the restaurant at 4 p.m. and finished their meals before the saleswoman and her friend entered the establishment at 5:15 p.m. The two groups were seated about 21 feet apart in the restaurant, or more than three times the recommended spacing between tables in the United States. The student left the restaurant five minutes after the saleswoman entered. The student and the saleswoman had no interactions inside the restaurant, the study notes.
Diner C entered the restaurant at 5:22 with two companions. They were seated nearly 16 feet from the infected saleswoman. The two groups of diners overlapped for 21 minutes before the saleswoman and her guest left. The authors suggest the high school student and diner C were infected — and not their companions — because they directly faced the air flow circulated from the saleswoman’s table. Other customers outside the air flow from the saleswoman’s table also escaped without contracting the virus, even though they had spent more time in the restaurant with the infected diner.
Understanding the role that air flow plays in transporting droplets and aerosols large enough to infect others is important, scientists say. But such research also reinforces the randomness of infection as well as the limitations of the current protections that people use to keep themselves safe.
“There’s a degree of chance here because you don’t necessarily know where you’re going to be seated, and you don’t know what the air flow in the restaurant is. So you’re taking a chance when you go in,” says Martinello, the Yale associate professor.
Martinello said there was important information missing from the study, including whether the AC units had filters to clean the recirculated air inside the restaurant. What’s more, he added, the researchers didn’t publish the genome sequences that they had analyzed, which was a red flag. This particular novel coronavirus has a “fair amount of genetic material," Martinello points out.
“They say they’re the same,” Marintello says about the genome sequences, which supports the idea that the saleswoman likely infected the others. “But were they exactly the same? What was that degree of relatedness? And how did they define that they were the same? They don’t get into that, which I wish they did. That may be a reason why they ended up in this journal rather than something with a higher profile.”
Despite the flaws, Marintello and other scientists are generally on board with the study’s findings. They point out that the findings are similar to earlier research, including a study into a covid-19 outbreak tied to an air-conditioned restaurant in Guangzhou, China, even though the tables were much closer together in the Chinese case.
The study confirms, says Milton, that infected people “can shed virus into the air, and it can travel long distances, and the more air movement you got, the more you might have” the kind of scenario laid out in the South Korean restaurant study.
With more the 20 years of experience in his field, Milton doesn’t think it’ll be another 100 years before the United States experiences a pandemic similar to the one we’re living through now. As such, he would like to see governments and entrepreneurs invest in technology to make buildings and public spaces safer in the future. He says restaurants could install units now that would protect their customers: Upper-room ultra-violet germicidal systems, together with fans that push air to the top of indoor spaces, could be installed for anywhere between $1,500 to $2,000 per room, Milton says.
If the restaurant in South Korea had had ceiling fans above every table and upper room UV, Milton adds, the spread of the virus “could never have happened.”