I wrote something for Contexts magazine last month exploring the laggard US response to COVID-19. In it, I called attention to a March 1 story in the NY Times that reported that genetic typing of coronavirus cases suggested the disease had been circulating in Washington state for weeks, which was troubling because it underscored how badly the US response had fallen down. As of today (April 23), we now know that the earliest known death of coronavirus in the US was February 6. This substantiates what many had feared and means that spread was taking place for much longer than had been established. More generally, it underscores how much we still don't know about the true spread of the virus, the overall caseload, and how much work we have to do to determine credible answers.

As journals and presses have shown growing interest in work on COVID-19, even going to the unusual lengths of offering expedited reviews of work on COVID-19, I have been thinking a lot about what this means for the production of comparative historical knowledge. To be clear, I am very glad to see interest in academic work by journals and publishers on COVID, and in particular work by social scientists on the politics of national responses. More of this kind of work is needed, and not just on COVID response, but on the politics of health policy more generally. However, I am also acutely aware of the challenges involved in case comparison even when the statistics we do have are solid.

A recent news article suggested Brazil's official case totals were underestimated by a factor of 12-fold. China recently revised its death toll in Wuhan by exactly 50%. When we cannot rely on the numbers that offer a window into nations' responses, how do we compare responses?