At the daily Covid-19 briefings, the federal administration in Washington D.C. is referencing modeling by the Institute for Health Metrics and Evaluation (IHME) regarding the number of Covid-19 deaths in the USA. IHME’s modeling (here) projects that the number of daily deaths peaked yesterday, Friday April 10th. I think that is unlikely. Here’s why.
Figure 1 plots (dots) the day/day ratio of the number of total deaths in the USA for the last two weeks. Data is from worldometers.info. Including full weeks is important because the reporting each day appears to be sensitive to the day of the week (here). The daily growth rate is declining. Also shown as a solid line is the best fit (squared error) reduction line. The day/day growth rate is declining 94.4%/day.
Figure 1 Day/Day Ratio of Total Covid-19 Deaths in the USA Over the Last 2 Weeks
As of yesterday, there were 18,747 total deaths and 2,056 new deaths, a 12.3% increase from the day before. If yesterday was the peak, there would be 2,056 or fewer deaths today (Saturday). The day/day growth rate would be 11.0% or less, which would be an at least 89.4% reduction from yesterday. 89.4% would represent a sudden and substantial change from 94.4%, which is the best fit over the last two weeks. It is possible that today’s number is below 2,056 because reporting on Saturday’s and Sunday’s tends to be light (again here), but then I would expect to see the numbers “catch-up” early next week.
Figure 2 shows the projection for the number of daily deaths through the rest of April. Data through 4/8, 4/9 and 4/10 are shown to indicate variability. The declination rates for these three days are 94.6%, 93.9% and 94.4%, respectively. The peak days range from 4/21 to 4/25.
Figure 2 Projected Number of Daily Deaths in the USA
I certainly hope we have already reached the peak, but if we have not, the peak looks like it will be later this month.