What's Going Down?

Paranoia strikes deep
Into your life it will creep
It starts when you're always afraid
Step out of line
The man comes and takes you away
--Buffalo Springfield

Each year the Center for Disease Control (CDC) estimates the number of deaths due to seasonal flu. Here are the results for the flu season ending spring of:

2011: 36,656
2012: 12,447
2013: 42,570
2014: 37,970
2015: 51,376
2016: 22,705
2017: 38,230
2018: 61,099
2019: 34,157

avg: 37,463

These numbers are estimates, however, and CDC acknowledges the possibility of large error. Its own confidence intervals assigned to these values range broadly. For example, for the highest flu year in the series, 2018, CDC estimates that the true value of flu deaths could have been as high as 95,000. That's more than 50% higher the 61,099 point estimate.

These numbers are going to get more attention as COVID-19 mortality forecasts continue to ratchet lower. The University of Washington's Institute for Health Metrics and Evaluation (UW-IHME) projection, which has gained (for reasons currently beyond my comprehension) social status as a model of legitimacy, has been steadily revising its death forecasts (along with its forecasts of hospital resource capacity utilization) lower. As of this writing, it now projects nearly 82,000 domestic COVID-19 deaths within a confidence interval of approximately 49,000 to 136,000.

What this means is that revised COVID-19 death forecasts are now overlapping ranges associated with common flu mortality. With current US fatalities attributed to COVID-19 standing at about 11,000 and daily new case and death counts breaking lower, more downward revisions seem imminent. Only a couple of days old, the most UW-IHME forecast of about 14,600 deaths by today (~30% higher than actual) suggests these folks will be heading back to the drawing board real soon.

As COVID-19 projections continue their descent into the seasonal flu zone, more people will question what's going on down here.