92770239 - tired passenger sleeping on the airplane at window seat

Irrational, or rational, people are afraid of getting on planes right now. Some think all travel is bad, while others don’t mind it, provided travel is done safely and responsibly, between two areas with low infection rates and adequate medical facilities. But in both camps, fears around the actual journey still exists. Should it?

A new study from top scientists at MIT has greater insights than ever before into the likelihood of catching covid-19 on a flight, and how travellers are still making their way around the world, even on planes where people may have covid-19, without catching it.

According to MIT Professor Arnold Barnett, the chances of catching covid-19 on a domestic US flight with a duration of two hours, are 1 in 4,300, or 1 in 7,700 if the airline blocks middle seats. The new study caveats results with a margin of error factor of 2.5, but the number are still compelling, if true.

These death-risk levels are considerably higher than those associated with plane crashes but comparable to those arising from two hours of everyday activities during the pandemic

Arnold Barnett, MIT
a person holding test tubes

The purpose of the study was simple: to settle the debate between whether blocking middle seats actually mattered.

The science behind whether blocking middle seats can make an impact has been highly contested, with top officials such as Dr. Anthony Fauci expressing dismay at airlines, including American Airlines, for refusing to block middle seats, while some say it doesn’t do enough to justify the financial impacts.

According to the results, which are awaiting peer review, blocking middle seats created a safety factor of 1.8, which is a huge margin. In layman’s terms, you’re just under half as likely to catch covid-19 on a flight if the middle seat is blocked, compared to flights where middle seats are not.

It’s the difference between the probabilities 1/4300 without middle seats blocked and 1/7700 if the middle seat remains blocked.

Methodology Behind The Study

The study examined three questions, which even to the non scientific community, seem rather logical. First, the study examined the probability that someone on board may already be infected with covid-19, which particularly in the USA, where probability was determined, is higher than elsewhere.

Second, the study aimed to quantify how successful universal mask and other hygiene and distancing measures could be in curbing the flow of covid-19, in the event of an infected traveller on board.

Finally, the study examined risks associated with where one might be sitting, in relation to the infected person during flight, and whether there was a difference in where the infected passenger was sitting – aka closer to the front of the cabin, middle, rear and so forth.

Sunset aerial view through airplane window over wings. Flying at sunset and looking out of the window and enjoying the panoramic view. Travel and transportation concepts

Air Filters Are No Joke

According to the research, high grade HEPA filters used on board aircraft, which don’t feature on common forms of mass transit are highly effective. The filters are said to make marked difference in stemming spread, even when there’s a covid-19 infected passenger on board, with the study stating…

We posit that the airlines are correct when they contend that the powerful air-purification systems on jet aircraft largely negate the risk of aerosol transmission of Covid-19. Thus, when a contagious passenger is in her seat in row 22, she poses little risk to another traveler seated in row 14.

Arnold Barnett, MIT

Going further, the study addressed the elephant in the room, which is the notion that you the passenger may be seated directly next to a passenger carrying covid-19, aka the spreader on the plane. It’s here that strong data science in favor of blocking middle seats is introduced.

A given passenger can get infected, however, by droplets from a contagious passenger in the same row. But the risk depends on the distance between the two passengers. The meta-analysis of more than 100 studies in The Lancet (Chu et al, 2020; [6]) yields the approximation that infection risk is about 13% given physical contact with the contagious person, and that it falls by essentially a factor of two as the distance from that person increases by one meter

Covid-19 Spread Study

As expected, you’re most at risk if you’re literally rubbing elbows with an infected passenger, but with droplet transmission curbed through universal mask wearing, and distance creating an extra barrier, each meter cuts the risks in half.

Would New Seating Solutions Work?

The study surmises that adding plexiglass between each row, a concept which has already actually been introduced by cabin manufacturer Priestmangoode, could be of some benefit to airlines in the fight against covid-19.

a row of seats in a plane

Plexiglass barriers between each row are loosely estimated to be 1/4 more effective in blocking spread, though that’s not as wowing as it sounds. That’s because the study estimates standard seat backs to eliminate 3/4 of spread between rows.

It’s not hard to then understand why rigorous aircraft cleaning in between each flight can make meaningful difference.

Is Flying Worth The Risk?

If1/4,300 sounds a bit too close for comfort, the odds of actually dying from catching covid-19 are estimated to be much lower, somewhere between 1/400,000 to 600,000, according to the study.

The odds aren’t as unlikely as being involved in a fatal plane crash, which sit at 1 in 34,000,000, but are far better than the 1/2618 probability of dying from choking on food, the 1/104 chance of dying in a car crash, or 1/118,000 chance of dying from a dog attack.

The study gives nuance the notion that someone catching covid-19 on a plane can go on to cause harm to others outside of the flight, but these scenarios are duly modelled into the results.

Using the USA as an example also makes the study far from a pie in the sky “best case” analysis, given the higher rates of covid-19 transmission in the USA, compared with other countries currently open for travel.

Travel certainly isn’t without risks, or responsibility these days, but the study results, paired with real life data from countries like Australia, where sick people were on board, but no one caught covid-19 signal potential hope for a rebound in travel as practical measures and new treatments mount.

Gilbert Ott

Gilbert Ott is an ever curious traveler and one of the world's leading travel experts. His adventures take him all over the globe, often spanning over 200,000 miles a year and his travel exploits are regularly...

Join the Conversation


  1. Better check your math on those odds of dying from catching Covid-19. About 150,000 have died from the virus in the US. If about 1 in 500,000 dies from catching the virus, that would imply that the US has 75 billion people, or more than 10 times the population of the entire Earth.

    1. It’s the odds of catching it on the plane trip alone. There are many other places to catch it hence the discrepancy you raise.

  2. I think this is good information to use as a generalization that you tailor to your situation. I live in a small city that been designated a hot spot by the CDC and has primarily regional jets for commercial flying. These regional jets don’t have a middle seat to provide spacing and generally don’t have HEPA filters. This combination of circumstances tells me that I would likely have a much higher chance of catching COVID flying from my hometown, which changes my perspective on flying from pretty unlikely to hell no unless in dire need. Anyway that’s a personal example of using the information as a starting point. Thanks, more knowledge is good.

    1. Excellent point. I think only some airlines are blocking off regional jet aisle seats, so worth confirming with this list (below). Passengers definitely should take this time to learn more about the fleets their preferred carriers are operating, and in the case of international travel, etc, give preference to newer birds with the latest filtration and air circulation. Thanks as always, Christian!


  3. As has been pointed out out. It’s not JUST the plane where one might catch it. At the departing airport and and also at the arriving airport. Also factor in that you may get it and not be affected but now you are a carrier and could pass it on to someone in the high risk group. no simple solution for this problem.

    1. I totally agree with your comment here. These studies do not take into account, the train or cab ride to the airport, walking through the airport, security, the restrooms, the cafes, the boarding area where people are crowded next to each other (really there is no way or point to even try and prevent that), then the arrival airport, the jetway, the terminal, the shuttles, the rental car people (or cab/bus/train), then hotel staff, restaurant staff, etc. How many potentially hundreds of people will be in contact with that one single person who might have contracted aboard the plane?

  4. Of note, the study, on lines 335-338, assumes that the probability of infection over a two-hour period not flying is 1/5900. That is still a factor of 1.37 of flying vs staying on the ground (i.e., 37%)

  5. Has anyone else ever boarded a plane and the AC was useless while at the gate or is it just me? From time to time the ground equipment fails, is unavailable, or the aircraft’s auxiliary power units can’t meet the demands of the air conditioning system. It doesn’t happen often but when it does, it’s miserable. And in today’s COVID-19 environment it’s dangerous. All of the calculations I seen on the COVID risk while flying ignore the fact that the AC might not work until the plane is airborne. When the AC fails at the gate what do THOSE numbers look like?

  6. A 2 hour flight is a good start in the research, but I’d also like to know how/if the risk increases with the flight duration. With longer flights, more people get out of their seats and move around – bathrooms, standing, walking restless toddlers; plus, it is just prolonging the time in close quarters. I hope they will follow up with this analysis.

    1. Yes 2 hour flights don’t help explain much in the big picture . We can probably drive that distance. But a 5 hour or even transatlantic flight to visit family is concerning to me . Any thoughts on this?

  7. The odds could be 1 in 100,000 or more. If that one person is you or a loved one, statistics mean nothing – it directly affects you and that changes things. I might sound a bit ‘judgey’ here, but I think it’s careless, irresponsible, self-centered, and selfish to travel at this time. My two cents…

    1. Chad, I think there’s heartfelt sentiment in what you say, but I also think that statistics are one of our few abilities as humans to better understand, quantify and benefit from risk. I found this quote quite accurate…

      “Human beings, we just are not good at estimating our own risk,” said Ken Kolosh, manager of statistics at the National Safety Council, who oversaw the report. “We tend to fixate or focus on the rare, startling event, like a plane crash or a major flood or a natural disaster, but in reality, when you look at the numbers, the everyday risks that we face and have become so accustomed to form a much greater hazard.”


  8. Not so worried about my risk factor. Couldn’t accept unknowingly passing a death sentence to someone I care about. As much as I want and need to take a trip I will wait until there is a vaccine before taking to the sky.

Leave a comment

Your email address will not be published. Required fields are marked *