This animated map shows the estimated weekly incidence rate of COVID-19 by county in the United States. The rate for each date was calculated by summing the total number of new cases that occurred between four and ten days ahead of the mapped date, standardized by the US Census population estimate for 2019.
For example, the county rates reported for March 1st are calculated by summing the total number of *new* cases reported for that county between March 5th and March 13th, and the dividing that sum by the total population of each county. The lag time was estimated from reports on the time between initial infection and the onset of symptoms, based on existing reports (e.g., https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/symptoms.html). The idea is to provide a running estimate of the rate of new cases that occurred at a point of time, taking the lag time between initial infection and the confirmation of cases into account.
Each day in the animation lasts for about 1.6 seconds. The animation covers a total of 46 days, from March 1 to April 15, 2020. The “time scale” is therefore about 1:53,000 – one second in the animation represents around 15 hours in real time.
The rates are quite a bit more variable than the animation of cumulative confirmed cases, capturing the variable nature and quality of testing for COVID-19. You might notice that it is possible to see some stark differences across state lines, such as between Louisiana and Arkansas that begins to appear around March 15. This is pretty clear indication of different reporting rates and standards between the two states.
One potential application of this sort of mapping is to explore the impacts on varied state and regional infection control policies, but until testing data improves substantially, it will be difficult to use this kind of method for much more than a highly generalized and rough estimate of incidence rates.
Data for this map were downloaded from usafacts.org: https://usafacts.org/visualizations/coronavirus-covid-19-spread-map/. The original data were collected from the CDC and public health agencies, which are listed on USAfact.org’s download page noted above.
I produced this map with ArcGIS pro, version 2.4.3, Microsoft Excel (to prepare the data and perform the moving average calculations), and Adobe Illustrator.