Thoughts on COVID

We continue to try and sort out the data and information coming in on COVID.  At points it is a challenge to make sense of- but luckily groups like Data Driven Detroit are here to help.  See their page on how to interpret much of this data. 

If you are focused on incidence rates, I've been contributing to Drawing Detroit's analyses on the region. There are many decisions and assumptions that have to be made because of data quality, but we all need to identify those as best we can.

One issue we all have right now is how to separate out new confirmed cases from the fact that testing rates are also increasing.  A seemingly obvious way to do this though is to look at trends in newly confirmed cases as well as trends in deaths per confirmed case.  When we have reporting on total tests (both positive and negative this is not as big of an issue).  If these are truly new cases then the death toll should continue to increase with some lag; however, if this is a result of testing we should not see such an increase.  This is a testable proposition and we can determine the lag from new confirmed case to death and use this to understand whether increases in testing are the primary driver of newly identified cases.   The lack of testing has increased the death per confirmed cases in MI considerably according to Deadline Detroit.