What are your chances of getting sick on a plane? A new study offers mixed answers.


Can a new study tell you how to avoid getting sick on a plane—or just the best bathroom to use?

If you want to avoid catching a virus on a plane, take the window seat, avoid using the bathroom, and limit your interactions with the crew. Those are some of the take-home points from a new study that attempts to model how illness could spread around an aircraft. But because almost no sick passengers were part of the experiments, critics say you should take the new recommendations with a grain of salt.

The study provides a “brilliant” blueprint for how a virus could move around a plane, says Ira Longini, a biostatistician at the University of Florida in Gainesville. But, he adds, “I don’t think it’s of tremendous epidemiologic interest.”

Biomathematician Howard Weiss of the Georgia Institute of Technology in Atlanta and a team of about 10 graduate students and postdocs set out on 10 different transcontinental flights ranging from 3.5 to 5 hours long. Using an iPad, the scientists logged the movements of every passenger and flight attendant on board.

Unfortunately (or fortunately, for the passengers), the observed flights appeared to be filled with mostly healthy people, despite the fact that the data were collected during the North American flu season: “We observed about 1500 passengers on the flights and only one of the passengers was coughing moderately,” Weiss says. The researchers were also unable to detect any sign of viral genetic material in an analysis of 228 swabs taken from around the planes.

The paper provides a blueprint for the transmission network—an outline of which passengers are connected and how. But without an understanding of how likely illnesses are to hop from one person to the next, the study cannot show how the virus moves through the network, Longini says. It’s like having a road map, but no information on how far or fast the cars can drive.

To get around this, the researchers fed their model with transmission rates inferred from historical data—specifically one instance from 1977 in which 38 of 54 passengers and crew contracted an “influenzalike illness” while their plane sat on the tarmac for almost 5 hours with no air circulation. Then, they multiplied that rate times four to make it “kind of a worst-case scenario,” says first author Vicki Hertzberg at Emory University in Atlanta. With this high transmission rate, an infected passenger will cause 0.7 new infections per flight, the researchers write today in the Proceedings of the National Academy of Sciences.

However, Hertzberg notes that the decision to multiply the 1977 infection rate by four was “somewhat arbitrary” and that therefore the modeled number of new infections per flight is also arbitrary. “What we’ve shown is that there’s a perimeter of risk around the infection—one row ahead, one row behind, and two seats on either side—but beyond this perimeter, there’s very little risk,” she says. An infected crew member, however, could cause as many as 4.6 new infections per flight, even when using the fourfold lower rate of 0.0045 transmission per minute.

However, Longini says without better data on the actual transmission rate, this is basically a wild guess. But he also says the study might still contain some news you can use to avoid getting infected. The window seats were the most isolated from contact with other passengers by far. Predictably, passengers that got up and moved around had more contacts than those who stayed seated, and this trend was particularly pronounced for passengers seated midcabin. Lastly, if you need to go the lavatory—head to the back of the plane: The lines there tend to be only half as long.

The authors also stress that their model only applies to single-aisle flights ranging from 3.5 to 5 hours long. It also only considers direct, large droplet transmission of influenzalike illnesses. The way people move around on planes during longer or shorter flights or planes with other layouts will likely be quite different.

How much risk mitigation these strategies might confer is difficult to say—probably not much considering how low the risk of infection is unless you’re sitting very close to a sick individual to begin with. Neither Hertzberg, Weiss, or Longini said they would be changing their travel habits in light of the findings.