With the coronavirus spreading uncontrolled across nearly the entire country, many public health experts are urging Americans to cancel their Thanksgiving gathering plans.
But not everyone is taking this advice to heart: Roughly 40% plan to attend a Thanksgiving gathering with 10 or more people, according to a recent survey commissioned by Ohio State University's Wexner Medical Center.
Peer-reviewed risk assessment data produced by researchers at the Georgia Institute of Technology suggests that many of those big events may become coronavirus super-spreaders. Drawing on public data sets, the COVID-19 Event Risk Assessment Planning Tool produces daily county-level estimates of the odds of encountering at least one coronavirus-positive person at a gathering of 10 or more people.
At the county level nationwide, the average estimated risk of running into a coronavirus-positive person at a 10-person gathering is just a hair under 40%. That's a pretty high number — if you take five of next week's Thanksgiving gatherings, you can expect that a coronavirus-positive person will be at two of them.
This risk varies a lot by county. Coastal cities tend to be a little safer at the moment — the risk is generally between 10% and 20%, depending on where you are.
But in some parts of the country — particularly the Upper Midwest — the risk is much higher. In Cook County, Ill. (home to Chicago), the risk approaches 60%. If you attend two Thanksgiving dinners in Cook County, in other words, odds are you'll encounter a coronavirus-positive person at at least one of them.
In El Paso, the risk is nearly 90%. Across much of North Dakota, the odds are anywhere from 95% to 100%.
To calculate these risks, the researchers combine data on the number of people confirmed to be infected in a county, along with what's called the "ascertainment bias" — an estimate of how much confirmed case counts undercount the actual case numbers.
