We all die. COVID makes it sooner

We die one at a time. A tragedy and a cause to mourn.

A close look at mortality research and the big picture: "Demographic perspectives on the mortality of COVID-19 and other epidemics."

Age 93. High risk.

On a national scale, COVID just changes the mortality tables a little. Very little.

This post has the potential of offending readers, and not for the usual reasons. Trump-haters tell me they don't like it when I describe Trump's adroit messaging; progressive Biden-haters say they don't like it when I write that Biden did, after all, get lots more votes than Bernie Sanders; Biden-supporters don't like that I give Thad Guyer space to call Biden senile. Lots to hate here.

Today I risk offense because I acknowledge death. Worse, I "normalize" it by putting it into the context of other deaths for other reasons--background mortality--and describe it from the point of view of actuaries and demographic researchers, as statistics, not as human tragedies involving loved ones.

The key factor about COVID is that is that it primarily kills older, unwell people and usually spares the young.

This blog post is drawn from a research paper published by the Proceedings of the National Academy of Science of the USA, a paper written by Joshua Goldstein and Ronald Lee, both from the Department of Demography at University of California, Berkeley. Here is their paper: https://www.pnas.org/content/early/2020/08/19/2006392117

Demgraphers think big picture. The big picture is that COVID may be shaping up to be approximately equivalent to the HIV/AIDS epidemic and the opioid epidemic in its overall affect on Americans' life expectancy generally, but it differs in that those two epidemics have played out over two or more decades, and this one is concentrated in a few months so far. It is not as significant as was the Spanish Flu. The Spanish Flu affected more people in a smaller population, and its effect was primarily on young people entering their most productive years.

Observations and insights from the paper:

1. Risk of death at any age across all developed countries increases by about 10% per year after age 30, compounding from a very low number and growing higher as one ages. The United States is somewhat of an outlier among developed countries, with an increase of 8.6% per year over people younger than 30 because we have an unusually high death rate among young people in America due to underserved populations accessing health care. In the US and elsewhere, however, the 10% increase per year after age 30 settles into an increase of 11% trend, because of COVID.

2. The net effect, if a new wave of infections brought us 1 million COVID deaths in 2020, would be to reduce the life expectancy of Americans by 2.9 years. That is significant but it is not a return to some remote otherworld of misery. That would mean a return to the life expectancies that were in place in 1995. At the current target of 250,000 deaths in 2020, Americans' life expectancy drops 0.84 years.

3. In a population of 330 million Americans, about 3 million of us die every year, a death rate of 9.1 per 1,000. Again, if a million of us were to die this year from COVID, that would increase to 12.1 per thousand. At current rates of about 180,000 COVID deaths so far this year, and 1,000 a day new ones, we might expect 250,000 COVID accelerated deaths, which means the figure increases to 9.9 per 1,000.  

4. About 70% of COVID deaths are from people age 70 and above. The normal percentage of deaths by that group would be 64%. Because COVID deaths are concentrated in the same place that deaths are normally, COVID doesn't really change the pattern of deaths. It just accelerates it. More older people dying than usual.

5. Among deaths of people ages 35-54, more men than women die, at a ratio of 1.44 to 1, but this gender discrepancy erodes to 1.12 to 1 by age 85+

6. A big picture way to think of COVID deaths as "temporary aging." At 1,000,000 deaths in 2020, again, a figure that would imply a major new wave of infections, the temporary age shift is 8.5 years. That means a 30 year old would have the normal risks of dying that year of a 38 year old, and an 80 year old would have the risk of dying that year of an 88 year old. Of course, the risk of dying at both 30 and 38 is absolutely small. Thirty year olds don't dread the mortality risks of being 38.  However, with the compounding of death risks, the absolute risk for an 80 year old are high and the 88 year old much, much higher. That is where the consequences of this virus concentrate. However, we are currently targeted at 250,000 deaths, not a million. At the current rate target of 250,000, the age shift is 2.9 years. Still, more 83 year olds die than do 80 year olds.

7. The Social Security Administration calculates that the American population of 330 million has an average of 45.8 years of life still to live--14.9 billion person-years. If a million people were to die early from COVID in 2020, that number would be reduced by 11.7 million years of life expectancy, i.e. the years taken away early by COVID.  Note that this reduces the overall American life expectancy by only about 1/1000th. 

Why so small? Because the people who do die skew old, and taken as a group only have 11.7 years of life expectancy lost, and there are few in the cohort of the elderly compared to the broader population, and fewer still who die. COVID has huge consequences for people who get sick from it and die, but looked at big picture on its effect on the total overall years of life still to be lived by Americans, it is almost invisible. And, again, these people dying are rarely the parents of young children or people in early or mid career, as was the Spanish Flu. It kills old people, people like me. 

Note: The research paper is dense, but filled with observations and data like these. Worth reading.

Policy implications

Politicians of both parties have called this a war on COVID.

There may be lots more casualties than deaths. We don't know the possible health effects on people who get the virus and stay alive. 

Looked at as a whole nation, and recognizing that people are being born and dying all the time, COVID could be viewed as simply one of the "costs of doing business", unfortunate but inevitable in a large country in the real world. Orchardists expect insect damage, retailers some shrinkage. Donald Trump attempted that attitude early on, abandoned it, then returned to it. The cure is worse than the disease, he said.  A few people die and "it is what it is" he said recently.

Democrats make an issue of the death count. Trump has led GOP voters to treat COVID as a manageable, inevitable problem, like the flu or high blood pressure, something to live with.

"It is what it is" sounded cold and un-empathetic to many Americans, and Democrats have jumped on it. Trump is, in fact, singularly un-empathetic. He was thinking about his popularity and re-election, not the lives of others.  Being coldly rational about people dying is not what a president is supposed to say aloud, but it is the way a general fighting a war needs to think. Commanders told troops landing at Normandy in 1944 that they were heroes and that they expected 2% of them to die, now go in and serve your country. The army ordered a million body bags in preparation for the invasion of Japan in 1945.

Trump's behavior is not irrational, and indeed it may be too rational for the circumstances. Selling you the benefit of your death on behalf of the economy and his re-election may be beyond  even Trump's ability. Still, from a broad, national perspective, the 250,000 deaths don't change the overall years left to live very much.

As Bob Warren wrote here on March 10, to the alarm and angry disapproval of commenters, older people had their turn and might serve everybody by getting out of the way. Click: Time for us old people to get out of the way  No one wants to hear that. I certainly don't.

Seniors aren't volunteering, but COVID may be drafting them. There are a lot of draft resisters, and they vote.