Deaths as Reported by Minnesota Versus Date of Death by CDC

It has been a while since I did this analysis, mid-August or so.  The state of Minnesota reports deaths by date reported, which has nothing to do with when the person died.  But the CDC, thank God, reports them by state by week of death.  The chart below shows by week, the number of deaths as reported by Minnesota and as listed by the CDC for the same week.  I have pulled this data several times and the columns in the table show how the weeks fill out over time, and just how far back changes are still occurring.  Why I don’t know.  Even more bizarrely, although the state uses date of report, for the week ending June 20, the state decided that it actually reported 90 deaths instead of 89 when I looked this time.  I had assumed the date of report numbers would never change.

If you look at the table and the graph, you can see that in general the CDC’s actual date of death version has the epidemic peaking slightly higher and sooner than the date of report used by Minnesota.  July flattens for both, then it rises slightly in August for the CDC and Minnesota is beginning to report a higher number of deaths, many of which are from weeks ago.  The CDC data for the last month is still probably not complete, although the way Minnesota is reporting it could be months before we know the actual number by date of death.

Now look at the table and compare the CDC’s death by week report that was pulled on July 10 with the one I pulled on August 14 and the one from today.  Even from August to now, for the week of April 25, two deaths were added.  Seven deaths were added to weeks in May.  Why?  How can deaths still be added over four months later?  It is hard not to be suspicious that every effort is being made to attribute deaths to CV-19.  I will do this in another month or so and we will see how many more deaths are being added that supposedly occurred weeks ago.

The CDC has received death certificates for a total of 2129 deaths.  Minnesota is currently as of today listing 2212 deaths reported, and 53 that I can’t tell for sure if they are in that 2212 or not.  If it is 2212 than we have 83 deaths from the state that will eventually make their way to the CDC and be put in the right week, since the state won’t do that for us.  The other thing I learned from this is that the chart in the weekly report that is labeled deaths by week is not a report of date of death.  Even though there is no matching table, you can tell from the chart that it isn’t the same as the CDC report.