There Is Little Evidence That Mask Mandates Had an Important Impact During the Omicron Surge

Case trends in states with mandates were very similar to case trends in states without them.

As of yesterday, The New York Times reports, every state that required people to wear face masks in public places and/or in schools during this winter's COVID-19 surge has decided to lift those mandates, with the exception of Hawaii. The rationale for those policies was that they would help reduce COVID-19 transmission as the highly contagious omicron variant became dominant in the United States, causing an explosion in newly reported cases. Explaining the justification for lifting mask requirements, governors and public health agencies generally have cited the precipitous decline in daily new cases since mid-January. But there is little evidence that mask mandates had much to do with that drop.

Nationwide, according to the New York Times database, the seven-day average of daily new cases rose sharply from mid-December to January 17 and has fallen just as sharply since then, which is consistent with the experience in other countries after omicron emerged. The same basic pattern was seen in nearly every state. Overall, there is no obvious difference between states with mask mandates and states without them.

Ashish Jha, dean of Brown University's School of Public Health, notes that the graph for California, which required masks in indoor public settings beginning on December 15, is "strikingly similar" to the graph for Florida, which had no mask mandates. He adds that the two states "had nearly identical infection rates."

That second observation requires a caveat, since testing rates affect the number of cases that are recorded. California tested a larger share of its population at the peak of its omicron surge than Florida did, and Florida's positivity rate was higher, which suggests that Florida missed a larger share of infections than California. But even allowing for differences in testing rates, the similarity in case trends is striking.

"At first blush," Jha says, it "looks like their different COVID strategies did not end up mattering much." But despite an official infection rate that was virtually the same as California's, he adds, "Florida had 33% more deaths per capita than California" during the omicron surge.

If Florida's case numbers understate the true number of infections more than California's do, that could partially explain the difference in COVID-19 deaths per capita. So could age demographics: The median age of Florida's population is 42.5, compared to 37 in California. Jha thinks vaccination rates are another important factor: While "only 59% of eligible Florida seniors have gotten boosted," he notes, the rate for California is 70 percent.

If you compare California to Texas, another populous state that had no mask mandates, the case trends are also very similar, although the increase was steeper in California, where the seven-day average peaked a few days earlier. The graph for New York, which also had a statewide mask mandate, looks very much like the graph for California, but with an earlier peak.

"Masks have been the most visible part of America's pandemic response, but one of the least consequential," science journalist Faye Flam writes in a Bloomberg Opinion essay. "The states with mask mandates haven't fared significantly better than the 35 states that didn't impose them during the omicron wave. Rhode Island, where I live, has had a mask mandate since mid-December; nonetheless, we saw our January surge rise far higher than any other state. There's little evidence that mask mandates are the primary reason the pandemic waves eventually fall—though much of the outrage over lifting mandates is based on that assumption. Many experts acknowledge that the rise and fall of waves is a bit of a mystery."

Laboratory studies indicate that masks, especially N95 respirators, can help reduce virus transmission. But as Flam notes, "the benefits of universal masking have been difficult to quantify" in the real world, where cloth models predominate and masks may not be clean, well-fitted, or worn properly.

The strongest real-world evidence in favor of general masking comes from a randomized trial in Bangladesh, which found that the use of surgical masks reduced symptomatic infections by 11 percent. That's not nothing, but it's a pretty modest effect, and it was achieved with surgical masks worn by adults in conditions that encouraged proper and consistent use. The same study found that cloth masks did not have a statistically significant effect.

The Bangladesh study, in other words, does not demonstrate the effectiveness of the cloth masks that adults tend to pull out of their pockets when they enter businesses, let alone the masks that children as young as 2 are forced to wear in schools and day care centers. If you've seen pictures of those kids, you know they are not necessarily using masks as intended.

Given the situation during the omicron surge, there are additional reasons to doubt that mask mandates, even with perfect compliance, had much of an impact. While mandates required shoppers to don masks before entering supermarkets, for example, the risk of transmission in such settings is low, given the amount of time customers usually spend in them, the size of the air space, and typically wide distances between patrons. Conditions in bars and restaurants are more conducive to virus transmission, since customers spend more time there in closer proximity to each other, often while talking. But since people were allowed to remove their masks while eating and drinking, requiring them to cover their faces upon entry was more a symbolic gesture than a serious safeguard.

Beyond the question of how effective masking is in practice, there is the question of what impact mask mandates have on behavior. Even if masking works, that does not necessarily mean mandates do.

An Annals of Epidemiology study published last May found that mask mandates in the United States were associated with lower transmission rates from June through September 2020. "The probability of becoming a rapid riser county was 43% lower among counties that had statewide mask mandates at reopening," the researchers reported. But the study did not take into account other policies or voluntary safeguards that may have differed between jurisdictions with and without mask mandates. Nor did it look at actual mask wearing, as opposed to legal requirements.

Based on data from various countries and U.S. states from May to September 2020, a preprint study published last June found that general mask wearing was associated with a reduction in virus transmission. But the researchers found no clear relationship between mask mandates and mask use. "We do not find evidence that mandating mask-wearing reduces transmission," the authors reported. "Our results suggest that mask-wearing is strongly affected by factors other than mandates."

An August 2021 systematic review of 21 observational studies found that all of them "reported SARS-CoV-2 benefits" from mask mandates "in terms of reductions in either the incidence, hospitalization, or mortality, or a combination of these outcomes." But "few studies assessed compliance to mask wearing policies or controlled for the possible influence of other preventive measures such as hand hygiene and physical distancing."

Like the debate about lockdowns, the debate about mask mandates will continue. Because there are so many variables to account for, it is very difficult to isolate the impact of any given policy. But it seems clear that anyone who takes it for granted that mask mandates have played a crucial role in controlling the spread of COVID-19 is making a series of assumptions that are not justified by the evidence.