Blog Archives

Image

US Mortality Projection by Age

Yesterday I did a rough estimate of the number of deaths in the US using worst case, middle, and best case scenarios. I think what might be more descriptive is to paint the picture for how this will unfold in different age groups.

For this analysis, I used the estimates of the US population for 2018 by the US Census Bureau and the case fatality rates (CFR) for 10-year age strata in China as reported by the China CDC. The overall CFR falls between my best case and midpoint projections yesterday.

Age, yearsDeaths
 0–9 – insufficient data
 10–1917,135
 20–2918,015
 30–3917,438
 40–4932,528
 50–59110,490
 60–69271,945
 70–79365,481
 ≥80369,209
Total1,202,243
  1. While this does obviously impact older populations much worse than other groups, it’s important to note that tragedy is going to be felt at every level.
  2. This is a good illustration of why I project nursing homes, assisted living facilities, and retirement communities are going to face a devastating event.

The most important thing though is to recognize that your behavior is going to impact the lives of others and those they love. If you get this virus, you can be spreading it for days before you show any signs or symptoms. I can’t express enough how important it is to STAY HOME UNLESS ABSOLUTELY NECESSARY. Don’t go to any gatherings. That means church, school, meetup groups…everything. It’s the only way we are going to win against this.

Shortly after writing this, I received a video from a friend. It’s in Italian, but it’s very easy to understand what is happening. The narrator is comparing the obituaries in his newspaper, the first on February 9th, “one page”, the second on March 13th, “ten pages.” If that doesn’t wake you up, I don’t know what will.

Image

500,000

I had done a little calculating in my head last night and was very disturbed. I wasn’t even sure if I should talk about it, but I decided I would sleep on it and run the numbers more carefully this morning. I had the same result.

I used a case attack rate of 20% (that’s the percentage of the population that will get infected). To put that number in perspective, the 1918 Spanish Flu is estimated to have infected 20-30% of the world population.

Next, I used the case fatality rates I’ve been tracking for weeks. For the low end, I used the numbers from South Korea, for the midrange the global average, and at the high end, Italy.

Without aggressive social interventions (school closings, banning any kind of mass gatherings, mandatory quarantine of entire cities, here’s the the number of deaths I’m projecting in the US using a very gross analysis of the data. I’ll be working through new models breaking this down by age categories this weekend.

Low: 506,896
Medium: 2,422,566
High: 4,377,735

In context, a normal influenza season in the US, we have about 36,000 deaths each year. Do you understand why this is different than influenza?

I don’t know what else to do to get people to take this seriously. We are in trouble and are way behind the eight ball in the US.