Flight attendant coronavirus risk
Just this week United Airlines’ CEO, Scott Kirby, said the following:
“At United, but also at our large competitors, our flight attendants have lower COVID infection rates than the general population, which is one of multiple data points that speaks to the safety on board airplanes.”
Then Delta Air Lines’ CEO, Ed Bastian, said the following:
“If the experience of flying was not safe, you’d expect our people to get sick. We track the health of our people. Our people are meaningfully less infected than the general population.”
Here’s the data used to support this claim:
- There are around 122,000 people in the US employed as flight attendants, and just over 1,000 have tested positive for coronavirus, for a 0.8% incidence
- There are 330 million people in the US, and there have been 6.6 million coronavirus cases, for a 2% incidence
That suggests flight attendants are getting coronavirus at only around 40% the rate of the general population.
As Sara Nelson, the President of the Association of Flight Attendants, describes these statistics:
“I really want to applaud our airlines and our airports who have really stepped up. I think all of that is evidence that the policies that have been put in place and the practices that have been put in place have helped to really decrease the risk of spreading coronavirus and in a lot of ways really control it in air travel better than on almost any other place in our communities.”
What could this be attributed to?
First of all, I think it’s worth acknowledging that this isn’t a peer-reviewed study. Rather this is the internal data that airlines have on employees.
While there are 122,000 flight attendants in the US, many of them have been flying very limited schedules. While I imagine they’re supposed to report cases to the company, I also wouldn’t be surprised if not all cases were reported correctly, especially in situations where flight attendants didn’t have any flying on their schedule.
I still expect that this data is at least close to being accurate. This is also a good reminder that if you’re scared or cautious about traveling, it’s probably not the flight you should necessarily be concerned about.
When the pandemic started, the thought of getting on a plane sounded terrifying, since people don’t usually put “airplanes” and “great air quality” in the same sentence. But that perception is hopefully starting to change due to a few factors:
- The HEPA filters that airlines use are really good, and fully circulate air every few minutes; this is better than just about any other indoor space you’ll find yourself in
- As much as we hear stories suggesting otherwise, airlines do a better job enforcing mask policies than just about any other businesses where proper distancing isn’t possible
- When you consider that tens of millions of people have flown since the pandemic started, there are very few cases traced to planes
- Passengers should be most concerned about the risk of everything else when traveling, including airports (particularly security checkpoints, jet bridges, and bathrooms) and what they do at their destination
Based on that you may not be surprised to learn that flight attendants have average cases, but how do they have significantly below average cases? I’d speculate that it probably comes down to mask awareness, even beyond the job:
- Airline employees have been doing a phenomenal job wearing masks correctly, in my experience, much more so than those in any other line of work
- Flight attendants understand the importance of masks, and probably take those precautions even when not flying, since they get so used to it
Data shared by airlines and unions suggests that flight attendants are testing positive for coronavirus at a rate that’s less than half of that of the general population. While I wouldn’t assume the data is 100% accurate, I do think it’s generally true.
That further supports that flying with proper precautions is fairly low risk, and it probably also reflects that flight attendants are more likely than others to follow best practices even when not working.
What do you make of this data?