Heading out? Assess your COVID risk
Heading out? You might feel safe even after days of record positive test rates for COVID-19 in Connecticut.
But maybe you want more than a feeling to usher out the second calendar year of coronavirus. You want a way to assess the risk — beyond knowing that one of every seven people tested over the last week came out positive, and that the number of people in Connecticut hospitals with COVID grew by 314 in just seven days, with 40 percent of them fully vaccinated.
Let’s say you’re considering joining a party or eating in a restaurant with 25 people in the room. What’s the chance that at least one person has an active, positive case of COVID?
It’s 74 percent in Fairfield County, 73 percent in New Haven County and — breathe easy — 66 percent in Hartford County, according to an intractive tool developed by researchers at the Georgia Institute of Technology.
If the crowd is smaller — 15 people — the chance that one of them has active COVID-19 is 56 percent, 54 percent and 47 percent in Fairfield, New Haven and Hartford counties, according to the model, which is updated daily based on real-time data and illustrated on a colored map showing every county in the nation.
Accurate? Meh. Experts I spoke with see all sorts of logical flaws in this kind of glitzy calculation. It doesn’t account for hot spots within a county, the current strains, whether people in the gathering are vaccinated, how many got tested before they went out, and on and on.
“You have to take all these things with a grain of salt,” said Dr. Albert Ko, professor of epidemiology at the Yale School of Public Health and co-chairman of Gov. Ned Lamont’s 2020 advisory committee on reopening the state.
“These type of models don’t do well when you’re on the upswing or you’re on the downswing,” Ko said — and yeah, we are most definitely on the upswing here in the Northeast.
Still, the “COVID-19 Event Risk Tool” as it’s known, has value as an intuitive eye-opener, as Ko put it. It’s a different way of thinking about risk besides the big population numbers we see every day. And it’s kind of fun.
On Tuesday, before I knew about this Georgia Tech analytic, which rolled out in the summer of 2020, I was thinking about how people partying over the weekend might assess risk. My daughter, her boyfriend and I ate outside, a balmy 45-degree day, at a West Hartford restaurant.
As I walked past packed tables to the bathroom in the back, I wondered whether anyone had an active case of the illness. The Georgia Tech website, I later learned, said it was 88 percent likely based on a headcount of 50, and none, of course, wore masks at their
tables.
Like most people, my friend Sharon Franklin in Stamford guessed lower percentages than what the Event Risk Tool showed. But her own risk assessment caused her to stop going to restaurants after last Sunday’s brunch — even before the shocking spike in numbers this week — and she canceled a small New Year’s Eve dinner party of four people at her house.
At work, where she’s managing director of campaign and community engagement at the United Jewish Federation of Stamford, New Canaan and Darien, Franklin and her small group of in-office colleagues had not been wearing masks around each other.
“Then this week, everything changed. I kept track of the numbers
and what was happening and I heard from friends and acquaintances that some people got COVID,” she told me.
Hearing fears from her daughters in New York City helped the assessment along, too, as stories of new cases, even among vaccinated people, piled up.
That’s a common-sense approach to assessment that combines numbers with intuition.
“I don’t think you have to be a rocket scientist to understand that, hey, we’re in the upswing of a very large epidemic. Many of us know many more people who have been infected with COVID than in previous waves,” Ko said.
More people infected but with less severe illness, at least that’s the hope with triple-vaccinations and the more contagious but less virulent omicron strain. Unlike in the spring of 2020, when we just shut down if we were lucky enough to work from home, we’re assessing risk a lot more closely as 2021 turns to 2022.
Even with alarming positive test rates, “I do think risk assessment is still a factor in the sense that if you are vaccinated and boosted, your risk is far lower of going to the hospital than if you’re unvaccinated,” said Dr. Manisha Juthani, commissioner of the state Department of Public Health.
Juthani, associate professor of epidemiology at the Yale School of Medicine, takes a cautious but pragmatic view. “I know some people will make the choice, ‘Look, I’m vaccinated and boosted and I want to go out,’ ” she said, adding that it’s smart to wear a mask around people indoors — and even outdoors, “if you’re in a crowded, packed kind of environment.”
Unfortunately, Juthani’s and Lamont’s efforts to bring 3 million rapid, at-home tests to Connecticut last week fell through, eliminating an assessment tool for a lot of people. The perverse bright side: The COVID spike was so sharp that many people who would have wanted tests to go out are now just staying home anyway.
As for the Georgia Tech tool, Juthani, like Ko, sees large variables that could swing the numbers dramtically — notably whether people are vaccinated. And even if there are active cases, your chances of catching COVID depend on all sorts of factors.
It’s fine to look at how the numbers change over time — it’s getting better or it’s getting worse — but the danger of looking at slice-in-time risk numbers, Juthani, Ko and other experts said, is this: You might say, oh, it’s only a 20 percent chance, and then do something dumb, or end up exposed to a super-spreading sick person.
So Happy New Year, assess wisely and remember Ko’s words of wisdom for the start of 2022.
“We don’t really need a model to tell us what’s obvious already,” he said, “because we’re in the upswing of this epidemic.”