Coronavirus Recoveries
Many regions across the United States are currently experiencing a surge in new coronavirus cases. Curious readers may ask: How quickly are these new cases progressing?
As we’ve discussed before, local-level data may not address every question, and we’ve used data from Sangamon County, Illinois, as an example. We’ve grouped data provided by the Department of Public Health (scdph.org) into three statuses: New, Existing (Isolated plus Hospitalized), and Closed (either Recovered or Deceased).
The graph below shows total cases over time (through Nov. 23), broken out into these three groups. Since we are viewing cumulative cases, it’s easy to visualize the portion that have occurred just since the beginning of November.
We can also make a few more observations.
First, we can see that while the numbers of new cases naturally get lots of attention, there is some good news: Recoveries are also proceeding at a brisk pace. We see this from the rate that the Closed status is increasing. (‘Closed’ also includes fatalities, but as of this writing the recovery-to-fatality ratio is more than 54 to 1, so the Recovery totals are within 2% of the Closed values.)
Second, we know that recovery from the virus takes time, and so an increase of the rate of the ‘Closed’ status must necessarily lag any increase in the New status. This lag time is quite visible in the October/November time frame, as the ‘Closed’ rate of increase nearly mirrors the cumulative total from a couple weeks prior.
Penciling-in the rates highlights a visual similarity
On a related note, we see that the total cases Closed at the end of the the data nearly equals the cumulative total just before the sharp increase in New cases (early November). We can therefore hope to see a corresponding sharp uptick in the recovery rate, perhaps quite soon.
Due for a faster recovery rate? We’ll know soon.
We plan to update this graph as new information is available. The latest will be on our Sangamon County Coronavirus page.
Notes: Data for New cases is available on a daily basis, while other data has been inferred when unavailable (typically weekends), using New Existing = Previous Existing + Previous New, and New Closed = Previous Closed until the next update. This leads to a stair-step effect for the Closed data series, which is most apparent when the next update shows a substantial increase (for example, over weekends in November).