ABOVE: Ashok Srinivasan
UNIVERSITY OF WEST FLORIDA

In its latest guidance on air travel, the US Centers for Disease Control and Prevention has stated that flying domestically is safe for people who are fully vaccinated, but questions remain about the risks for those who aren’t immunized. Studies have suggested that the actual flight is fairly low-risk—even without masks, infectious droplets aren’t likely to spread past a couple rows, for instance—but lots of behaviors associated with flying, from security screenings to getting on and off the plane, can add to potential infection exposure. 

The Viral Infection Propagation Through Air-Travel (VIPRA) project brings together scientists from diverse fields to model and analyze these different behaviors and potential strategies for reducing the risks they pose, and in a paper published April 28 in Royal Society Open Science, they weigh in on the boarding process.  

The Scientist spoke with VIPRA’s Ashok Srinivasan, a computer scientist at the University West Florida, about the team’s findings.

The Scientist: What is known about the risk of catching COVID-19 from air travel?

Ashok Srinivasan: Well, we know for sure that it is not as high as people think it is. On the other hand, it is not as low as the airline industry tries to make it sound. There’s a big problem here in that there’s really no good contact tracing done on people who have actually flown in the air. So, we really don’t have very good information on what is actually happening. We have papers that people have published, but most of the incidents don’t get recorded in papers. So, you see a lot of cases where we have a few superspreading events, a lot of people got sick, but that is not a typical situation. . . .

The air that comes out of the plane is quite pure, so you don’t get infected from that, you get infected before the air has gone through the filtration system. And [the filtered air] is much cleaner than you would find in restaurants and all that, so the risk of infection from air travel is not as high as in many other places. But, of course, you have a bunch of people packed together. . . . Because [there are] a large number of flights, and a large number of people traveling, even though the probability of infection is low, you are going to have some big outbreaks. 

TS: And how much does the boarding process itself contribute to the risks related to air travel?

AS: So, it looks like the boarding process contributes around twenty to twenty-five percent to the number of cases. . . . We also compared the boarding versus in-flight movement, to go to the restroom and all that, and boarding is much more of a risk. Deplaning is also not as much of a risk as boarding. So, apart from people being seated next to someone who is actually infected, the biggest risk actually comes from the boarding process, not from the other aspects of the flight.

The risk of infection from air travel or any other situation can almost be eliminated by using an N95 mask or equivalent, so it really can be very safe.

TS: In your study, which boarding methods did you examine?

AS: We looked at many options. . . . In the paper itself, we reported only four different processes because there were too many to report, but these are the most insightful. One is random boarding, which is like having just one zone. The other was having six zones plus business class, with business class boarding first. And then we said back to front, and back to front with the business class boarding first. And then we looked at variations of these with middle seat empty, no overhead bin storage, and window boarding before aisle.

TS: And which one was the best?

AS: The best is to have random boarding—only one zone. And among the variations of that, of course, if you keep middle seat empty, that is much better. And if you don’t allow the storage of luggage, that is much better. If you board window before aisle, that is much better. These are good for all boarding processes.

TS: Is that just because of the speed of boarding? I know that previous research has found that windows-first and random are faster than back-to-front boarding.

AS: That is partly the reason. There are two primary mechanisms through which the disease spreads. One is if you have people who are seated close to each other for a long time. . . . If the boarding process is slow, they’re seated together longer. So that’s one aspect. The other aspect is, when they are actually seated for a long time, are they sitting close to each other. In the back to front, they have to be seated close to each other. If you have random boarding, they’re distributed throughout.

The other [primary mechanism] is when someone is storing luggage, people usually need to wait for them to clear the way before they can get in. So, if you have more zones, then people tend to have large clusters in a few locations. If you have random boarding, you have small clusters in many locations. And the number of interactions in a small area is roughly quadratic to the number of people. So, it’s better to have many small clusters rather than a few large clusters.

They looked at the most insignificant mechanism and tried to minimize it, and they were successful at that.

TS: Are current airline procedures in line or not in line with what you found reduces infection risk?

AS: They actually changed the procedures, typically, to increase risk. This  comes from CDC guidelines, so they have an excuse. The reason is that there are three mechanisms through which the spread happens. They focus on one mechanism: when people walk past someone who’s seated, how much of a risk is there from that. But, it turns out that the exposure is very small. So, they looked at the most insignificant mechanism and tried to minimize it, and they were successful at that . . . . That is the reason the procedure that they changed was actually worse than what they had earlier, and much worse than a random procedure would be.

TS: What other risks of air travel would you like to examine that you didn’t get a chance to here?

AS: Well, so this is one paper; we actually sent it for review about a year back, so the review process was very, very slow. Meanwhile, we have actually done other work. And, I think, really, the risk of infection from air travel or any other situation can almost be eliminated by using an N95 mask or equivalent, so it really can be very safe. The most important thing is to actually have an N95 mask, and to not remove it for eating food and all that in the plane. . . . If people can avoid stowing in the overhead bins, I think that would be a good thing. On the other hand, if people wear N95 masks, then most of this actually doesn’t matter.

Editor’s note: This interview was edited for brevity.