Did rigid enforcement ruin mask mandates? ("Inside" Inside Medicine)
After a conversation with Ian Bremmer, I worry we may have scolded our way out of mask adherence, and something is better than nothing.
Welcome to Inside Inside Medicine. This is Premium Content, available only to paying subscribers. Thanks for being a member of the Inside Inside Medicine community!
When the federal mask mandate on airplanes and other public transit was invalidated by a federal judge this week, I saw a tweet by fellow Bulletin writer Ian Bremmer that really ticked me off. Bremmer is a respected political scientist, an author, and a well-established public intellectual. You’ve probably seen him on television or in every major publication, from Foreign Affairs to The New York Times. He’s everywhere. I have always found him to be interesting and insightful. So, when he tweeted that he was “happy to see the end of the public transport mask mandate,” my blood pressure skyrocketed.
“Why am I happy to see the end of airplane mask mandate? Because it didn’t make sense. Still serving drinks and food, not enforcing standards for effective masks. It was performative,” Bremmer tweeted.
Though we don’t know each other, I politely reached out to him (as a fellow Bulletin.com writer), and we had a good conversation that I think highlights an important point on mask use that I’m sharing with you here. The main point: some masking is better than no masking. Little breaks in masking do not negate the whole exercise (despite what you may have read in the New York Times today). Just to be sure I got it right, I spoke to my Harvard colleague Dr. Joseph Allen today, a leading expert on air quality. I also checked in with Dr. Abraar Karan, a former colleague of mine now at Stanford who has a lot of expertise on masks. Joe and Abraar agree with my analysis on this one. Case closed.
Here’s an edited summary of what I wrote to Ian by email, and our subsequent correspondence (which he agreed to let me share).
•••
Ian,
The issue I am having with your tweet and interpretation of mask use is that it falls into the cognitive trap of letting “perfect be the enemy of the good.” Certainly, if everyone wore an N95 mask perfectly for a 5-hour flight, that would be better than 50%–75% adherence. I grant you that.
But catching Covid-19 is not like going out in the rain and getting wet (i.e., the second you get a drop of water on you, you are infected). The risk of contracting Covid-19 is more like going out in the rain with an umbrella. The fewer holes in the umbrella, the better the quality, the drier you stay. You get infected only if you get soaked. Or, maybe, you get infected only if coronavirus particles happen to land exactly on the tip of your tongue. No umbrella? The odds climb quickly. Leaky umbrella? It can still happen, but it’s a lot less likely. So, masks (and any intervention) should be thought of as lowering the odds of infection, rather than all-or-none.
We know that there are gradations of risk, even on planes. As I wrote earlier in Inside Medicine, one airplane outbreak study from March 2020—back when masks were optional—showed that the two things that were most highly linked to the risk of infection were proximity to the Covid-19 passenger and not wearing a mask. Non-mask use and being seated within two rows of the contagious person raised the odds of infection by 7 times each, in comparison to masked passengers seated in other parts of the cabin.
The cognitive leap that I think people have not made, two years into this pandemic, is that exposure does not equal infection. The risk of infection from an exposure is correlated to the time and intensity of that exposure. The intensity of exposure can vary greatly, depending on factors like the contagiousness of the source patient, whether you are wearing a mask, and the air circulation and purification in the environment. Why some exposures cause infections and others do not remains unclear, but we saw how inconsistent this can all be, as far back as the outbreak on the Diamond Princess cruise ship in February of 2020. Some spouses sharing a cabin managed not to infect each other for a few days, masked or not! Meanwhile, in superspreader conditions (which we barely understand), one source patient can easily infect dozens of people in a matter of minutes to hours.
Instead of looking at exposure as yes/no, we need to think in terms of gradients. It’s not “Did I inhale within a few feet of a Covid-19 patient, yes or no?” It’s “How many breaths did I take within a few feet of a Covid-19 patient?” and “Was I wearing a mask for most or all of those breaths?”
Let’s do a thought experiment that illustrates how wearing a mask most of the time still makes a huge difference. Imagine you’re flying and you are seated near a contagious Covid-19 patient. The odds of that are actually low (the virus is not that prevalent, thankfully), but that’s the “worst case scenario” that we need to ponder. Now let’s say that each breath you take carries a 1 in 3,000 risk of being “enough” to get you infected (I made this number up, but we need some kind of assumption to make this exercise work). If you take 5,400 breaths in a 5-hour flight (i.e., 18 breaths per minute for 300 minutes), the odds that you will be on the receiving end of an adequate “dose” of Covid-19 and will become infected is 85.5%. (For the math nerd, here’s how I calculated the probability: 1-(1-0.0003)^5400=0.855.)
Now let’s say that your N95/KN95 mask reduces your risk of infection by 95% per breath, and that you use your mask correctly for all but 20 minutes of the flight (while you eat). That’s 5,040 breaths masked, and 360 unmasked. Now your odds would be the sum of the risks while masked and the risks unmasked. By masking for most (93%), but not all, of the 5-hour flight, the risk of infection drops to 17.6%. That’s a lot better than 85.5%! If you unmasked for only 10 minutes, or took a shorter flight, the numbers get even better. In running these numbers, I found it interesting that in the second scenario, there was a 10.2% chance of infection during the brief unmasked time, and an 8.2% risk during the masked time. (Again, a math whiz will notice that these add up to more than 17.6%, but that’s due to the rules of probability, which I won’t go into here. Sorry/You’re welcome.)
You can also quickly see how a 10-hour flight in which you are perfect with your N95/KN95 mask would be superior to a 5-hour flight in which you took it off for meals. It all comes down to these little decisions. Alternatively, a 45-minute flight without a mask would carry a 22% risk of infection, while a 12-hour flight with perfect masking would carry a 20% risk of infection. Isn’t that interesting?
And this assumes one-way masking. If the Covid-carrying passenger was also masking for most of the time, your per-breath exposure would be far lower. So, maybe drop your mask and eat a little later than everyone else? The air turnover is great on planes.
As I wrote the other day, a loud minority of badly behaved anti-maskers made the airlines wary of enforcing mandates. Those poor flight attendants didn’t deserve being attacked. In hindsight, I wonder if demanding that mask-resistant people put their mask over their nose (not just over their mouth, let alone around their neck) created enemies who now will never wear a mask unless strictly enforced. On a plane of 300 people, asking 3 to fix their masks probably didn’t add much safety (other than establishing good group habits). But I wonder if, over time, we alienated those people and that, combined with toxic politics, we lost them for good by scolding them.
Given this thought experiment, I believe we can still derive tremendous utility from public transit mask mandates, even if adherence is imperfect, and even if we eat and drink on board.
Best,
Jeremy
•••
Ian replied to me, and largely favorably. I’m not sure whether I entirely convinced him (he didn’t delete those damn tweets). But in a subsequent reply, he asked whether I thought that, given what we now know, mandating and providing N95 or KN95 masks on planes would make sense (i.e., Why bother with lower quality masks?).
I responded that, in the pre-Delta and Omicron eras, it mattered less. But now, yes, mask mandates on airlines would be better if they specified (and even provided) good masks. But again, planes are relatively safe air environments (though we’re packed in like sardines, no?). If most people wore N95/KN95’s (and I recommend some of the newer, more comfortable options like the 3M Aura N95, which has been a real game-changer for me), and some only wore lower quality masks, the overall impact would still be substantial.
Ian agreed. “We have masks that work, they’re available, they’re comfortable…that should be the airplane mandate,” he told me. “Sounds like we pretty much agree on this.”
I think so!
•••
❓💡🗣️ What are your questions? Comments? Join the conversation below!
Follow me on Twitter, Instagram, and on Facebook and help me share accurate frontline medical information!