As protests against racism and police violence swept across the country, drawing massive crowds into the streets amid a pandemic, public health officials worried about what the overall impact would be.
Would these protests — which many health leaders said they support — also turn out to be virus super-spreading events?
But a new study by a nationwide research team that includes a University of Colorado Denver professor has found something surprising: The protests may have slowed the overall spread of the coronavirus in cities with large demonstrations, including Denver.
“We think that what’s going on is it’s the people who are not going to protest are staying away,” said Andrew Friedson, the CU-Denver professor who is one of the paper’s co-authors. “The overall effect for the entire city is more social distancing because people are avoiding the protests.”
“It’s nice to have some numbers”
Friedson’s specialty is economics — specifically the economics of health care. The field of COVID-19 research now contains a multitude of subspecialties, and it has often been economists leading the way in understanding how people are changing their behaviors in response to the pandemic.
As the protests built, Friedson said he and his colleagues took note of the rising concerns about virus’ spread. He said they also realized they had the ability to answer that question — using official coronavirus case counts and the anonymous, aggregated cell phone data that has become the gold standard for tracking societal shifts in movement.
The team worked quickly and published their findings earlier this month as a National Bureau of Economic Research working paper — meaning it has not yet been peer-reviewed.
“I’m someone who likes to get the answers out,” Friedson said. “There are a lot of people who say, ‘Well I think it should happen or I think this should happen,’ and it’s nice to have some numbers to inform these decision-making processes.”
Rising cases, rising worries
The paper comes as officials in Colorado and other states are concerned about rising infections, especially among young people.
New infections among young people have contributed significantly to Colorado’s uptick in cases in recent days — a rise that reversed a weeks-long trend of falling case numbers and has put Colorado back onto the list of potential coronavirus problem spots. Meanwhile, the number of new infections among older Coloradans has dropped.
With the July 4 holiday approaching, Gov. Jared Polis and county health officials have pleaded with people to be responsible and avoid large gatherings.
“We don’t have the direct causation of this uptick,” Polis told reporters last week, noting that there is evidence that some young people who are part of an outbreak in Boulder had attended protests while other outbreaks are tied to social gatherings. “And we hope this is a trend that is reversed in our state.”
On Monday, a spokesman for the Colorado Department of Public Health and Environment said that, while the state has now seen rising numbers of new cases for two consecutive weeks, “we have not seen any clear association between the protests and an increase in cases.”
The spokesman, Ian Dickson, said the uptick in infections “may be partly due to some Coloradans changing their behavior — especially socializing in larger groups, sometimes without proper distancing or mask wearing.”
Friedson said his paper doesn’t try to figure out whether the protests spread the virus among the people at the protest. Instead, he said the research took the bigger-picture view: What did the protests mean for overall transmission of the virus within the entire community?
The study looked at 315 American cities with populations of more than 100,000 and found that 281 of those cities saw protests. The remaining 34 cities that did not see protests — which, at the time, included Aurora — were used as a control group against which to measure the impact of the protests.
The researchers found that protests correlated with a net increase in overall stay-at-home behavior in cities where they occurred — and the increase was larger in cities that saw more sustained protests or reports of violence.
Not a green light
Friedson said he and his colleagues were a bit surprised at first. The protests in many cities, including Denver, were massive, drawing tens of thousands of people out to march. But they occurred in cities with hundreds of thousands to millions of residents.
“We started thinking about it a little more and we thought, ‘Oh my gosh we’re capturing everybody else,’” he said.
The paper also found that, with greater social distancing, COVID case growth slowed in cities with protests from what would be expected — but not by a statistically significant amount. There may be other explanations for the trends, the study’s authors note. Overall, though, they say the data show that any resurgence in coronavirus cases can’t be pinned entirely on the protests.
“Public speech and public health did not trade off against each other in this case,” the authors wrote in the paper.
But Friedson said there is one last important thing to keep in mind about this study: It’s not a green light for governments to fully reopen bars, concert venues and other places where people gather in large numbers. The key to the researcher’s conclusions is that the protests, while receiving lots of support, were ultimately things most people decided to avoid. That’s not true of many other large gatherings.
‘An outdoor wedding doesn’t generate avoidance behavior; we’re measuring avoidance behavior,” Friedson said. “People don’t say, ‘Oh man, there’s an outdoor wedding next door, we should stay home.’”