Multiple studies show virus transmission is possible on airplanes, but blocked middle seats don’t help

Last week American Airlines announced that it would book flights to 100% capacity starting on July 1st. Since then, the airline has faced a barrage of criticism from numerous sources, including, of course, Twitter.

Screen Shot 2020-06-30 at 8.56.09 PM

While these Tweets consistently rely on the “big company putting profits above safety” narrative, they also consistently lack any real basis for this accusation. While there are only a handful of studies on the transmission of viruses in a plane, all of these studies show that sitting next to an empty seat isn’t any safer. In fact, these studies consistently show that virus transmission on an aircraft can occur several rows away from an infected host.

An oft-cited study in the New England Journal of Medicine showed that viruses can indeed be transmitted aboard a plane. The study focused on a three-hour flight between HKG and PKG where a single host with SARS likely transmitted the virus to 18 other passengers.

Through thorough contract tracing, this study was able to create a seat map that shows infections relative to the host, below.

Screen Shot 2020-06-30 at 6.48.22 PM

What’s notable about the seat map is that the host (indicated by a “+”) was in a middle seat, but did not infect the passengers directly to his left or right; there would have been no difference in the infection rate if the host was surrounded by empty seats. What’s more troubling is that the host infected three of the six passengers in the row in front of him, and 14 passengers two or more rows away. There didn’t seem to be a “safe distance” from this host; more than a third of infections came in seats that were separated from the host by four or more rows.

This study does suggest that back-to-front transmission seems to be more likely, than front-to-back. 13 of 18 infections were in front of the host, while 4 infections were behind him. One was three seats to his left, separated by the aisle.

While one flight in the study did show 18 infections, two flights in the study, both with close to 300 passengers, showed no secondary infections despite the presence of one or more infected hosts. These flights were on larger 777 aircraft, compared to the smaller 737 with 18 infections.

Another study published by the US National Library of Medicine shows similar results for the transmission of the flu aboard aircraft. This study is unique in that it summarizes the results of 14 other peer-reviewed publications.

The table below shows its findings.

Screen Shot 2020-06-30 at 9.11.01 PM

Pay attention to the columns “Secondary Cases Identified” and “Secondary Cases Within 2 Rows”. You’ll see a similar, alarming trend. Of the 23 flights studied, 163 secondary cases were identified, and of these 68, or 42% were within two rows of the infected host, i.e. the “index case”. This means that 58% of infections occurred in seats three or more rows away from the host.

Some of American Airlines’ critics cite a Los Angeles Times article that states, “Health experts have told The Times that the risk of infection goes up as more passengers are crammed into a plane.” This is true for two reasons:

  1. A greater number of passengers means there is a higher likelihood that an infected host will be onboard
  2. A greater number of passengers means there are more passengers to infect

But should this increased risk alarm you? This is debatable, but you can do the math for yourself.

  • The US currently has 1.5 million active cases of COVID out of 328 million people, or an active infection rate of 0.46%. This means that out of a thousand people, there are likely to have slightly less than 5 infections.
  • An American Airlines 737 with 144 economy seats filled 85% full will have 122 passengers.
  • Therefore, on a flight that is 85% full, 0.56 passengers are likely to have COVID.
  • If the flights capacity is increased to 100%, there will be 144 passengers, and 0.66 passengers are likely to be infected.

Is there an increased risk? Yes. Is the 0.1 passenger increase something I would Tweet angrily about? Probably not.

You can take an even more extreme example like Alaska Airlines, which is blocking a third of its seats.

  • There are 147 economy seats on an Alaska Airlines 737-800, filled to 66% capacity you have 97 passengers.
  • 0.45 of these passengers are likely to have COVID based on the current 0.4% infection rate.
  • If the plane was filled to 100% capacity, there would be 147 passengers, and 0.67 passengers are likely to have COVID.

Again, there is a difference, but it is not significant.

What this data show is that we should not necessarily cling to empty middle seats as an effective way to prevent COVID transmission on planes. Transmissions actually appear to be more common in seats that are separated from an infected host by more than a row. And while decreasing capacity on planes will decrease the likelihood of transmission, the difference here is minimal.

I think the more important takeaway from this data is that flying on a plane is a risky activity, similar to a game of Russian Roulette. Once you get on that plane, whether it is 100% full, 85% full, or 66% full, there is a risk you get infected, and that risk is similar whether you’re sitting next to an empty seat or not. Therefore, after flying, passengers should exercise precautions like wearing masks, staying away from high-risk populations, and minimizing contact with large groups.

Yes, it’s fun to burn a multi-billion-dollar airline on Twitter. But passengers concerned about their safety and the safety of their loved ones should focus less on arbitrary systems of middle-seat blocking, and more on common sense methods of sanitation and separation post flight.