Question: The other day, I heard that 1 in every 100 Nashvillians is currently infected.  That got me to wondering, what proportion of Tennesseans have likely been infected to date?  

Answer:  To my knowledge, we don’t have a good *current* estimate.  As of September, our best estimate is that ~7% of Tennesseans had SARS-CoV-2 antibodies indicating previous infection.  
Recognizing that testing hasn’t captured everyone who has been infected, CDC is leading three types of seroprevalence surveys to estimate the proportion of the population that has been infected– large-scale geographic surveys, community-based surveys, and special population surveys.  Last month, JAMA published the latest results (September) from the large-scale survey, Estimated SARS-CoV-2 Seroprevalence in the US as of September 2020.  As of 24 September, researchers estimate that ~7% of Tennesseans had been infected.  You can see the seroprevalence estimates for Tennessee and other states in Figure 1.  To put things in perspective, after the height of New York’s initial wave, the survey estimated that ~23% of New Yorkers had antibodies. 

This survey has important limitations, a big one being that because it analyzes blood specimens collected for routine health screenings, its results may not be representative of the general population. Additionally, we’ve seen a huge wave of new cases since September, so these estimates are most surely out of date.  And since antibodies may wane over time, these surveys may underestimate rates of prior infections.  Finally, if you want to read more about seroprevalence in other countries as of September, this paper gives a nice synthesis– Population-based seroprevalence surveys of anti-SARS-CoV-2 antibody: An up-to-date review.

Figure 1. Seroprevalence estimates by state over time (JAMA)