D.C.’s Prevalence Estimates Are Pretty Accurate

Like many state* governments, D.C. releases a lot of data about COVID-19, including the number of identified daily new cases. Note the word identified. It’s never been entirely clear how well daily new cases reflect the number of infected people. Obviously, we can’t have fewer cases than reported, but is the number of reported daily new cases an underestimate of a few percent, or a huge amount (e.g., five-fold).

One piece of information that most governments are not releasing is hospital intake screening data, the results of people admitted to the hospital for non-COVID-related procedures, such as pregnancies or injuries. While these might not be a truly representative population sample, they are unbiased from the perspective of symptoms (most pregnant women are going to the hospital, and that’s obviously something that can’t be delayed). In lieu of that, we can try to find other estimates. One interesting non-governmental datum is this (boldface mine):

Ten Farmers Restaurant Group employees have tested positive for COVID-19 since the restaurant group behind Founding Farmers, Farmers Fishers Bakers, and Farmers & Distillers reopened its seven eateries to the public on May 29. City Paper doesn’t know this through investigative reporting. The restaurant publicly announced it to anyone else with access to the internet.

In July, Founding Farmers made its “COVID-19 Health & Safety Dashboard” available to the public. They began tracking data for internal purposes back in March. It looks like a Bingo board, but displays everything from the number of face masks the restaurant group has in stock to how frequently they clean and sanitize surfaces. The transparency stands out: Farmers Restaurant Group shares how many employees have been tested for COVID-19 and the results of those tests on the dashboard as well.

As of September 17, ten out of 755 employees had tested positive since May 29 (1.3%). In D.C. as a whole in the same period of time, 0.85% of D.C. residents have tested positive. I don’t want to make too much of this, and these are different populations: Founding Farmers employees likely don’t include children (though you never know…) or the elderly (or very few elderly anyway). Obviously, some Founding Farmers employees aren’t D.C. residents. And if Founding Farmers employees avoid getting sick, D.C. would ultimately pass there employee population. But it does seem that the daily new case data from May 29 onward are in the same ballpark as the Founding Farmers data. Given the low percent positive rate for D.C., this shouldn’t be a surprise, but it’s nice to see another data source that is roughly consistent with the daily case number.

Now, if only D.C. were releasing city-wide hospital screening data…

This entry was posted in COVID-19, DC. Bookmark the permalink.

1 Response to D.C.’s Prevalence Estimates Are Pretty Accurate

  1. peter says:

    It’s never been entirely clear how well daily new cases reflect the number of infected people. Obviously, we can’t have fewer cases than reported, but is the number of reported daily new cases an underestimate of a few percent, or a huge amount (e.g., five-fold).
    https://noithatdongthanh.vn/xuong-san-xuat-go-soi.html

Leave a Reply