Of all the statistical cacophony to which we have been subjected during the COVID pandemic, there is only one metric that really matters at the end of the day. How many more people are dying than normal? This is what epidemiologists refer to as “excess deaths.”
It is a fairly simple calculation. You compare the rate at which people were dying before the epidemic to the rate at which people were dying during it. The CDC has been tracking excess deaths from COVID in the US since the outbreak of the pandemic. For each week since January 2020, the CDC has run a calculation of the number of Americans who have died above what was typical for the same week in years before COVID.1
The great news is that it appears March may be the first month that the US could have fewer deaths than it did before the pandemic began.
In 2018 and 2019, an average of a little over 54,000 people died each week in the US. During the pandemic, that surged by more than 40% during the peaks of the three major waves in the US. For all of 2020 and 2021, the US averaged about 20% more deaths than normal. In the summer of 2021, before the Delta and Omicron waves, the excess fatalities got down into single digits but never fell below normal.
However, in recent weeks, the percentage has plummeted. The last week that the CDC shows any excess deaths was the week ending March 5, which currently stands at 4% above normal. The month of March is currently 17% below normal. Let me add the caveat that I have learned during the pandemic, that some states are extremely slow in reporting fatalities to the CDC. So, their numbers will grow over time, but the late data should not be enough to dramatically affect the numbers. My guess is that April will be the first month that Americans will die at a rate lower than before the pandemic began.
The CDC is still showing that a little under 400 people are dying each day from COVID, which is a little more than 5% of what was an average day before the pandemic. But the CDC reports fatalities based on the report date and not the date of death, so there is a significant lag time. Also, it is likely pneumonia caused by COVID will become a common cause of death among the elderly, displacing other types of pneumonia which have historically been a common cause of death among the elderly.
Assuming that there is not another deadly wave of COVID, which most experts discount at this point, there is a good chance the fatalities could run well below pre-pandemic levels for the next year or more. This might be the case because of a phenomenon known as mortality displacement. Mortality displacement occurs when an extraordinary event advances or delays fatalities that would have otherwise occurred at a different time. Typically, these events accelerate fatalities.
COVID was likely such an event because its death toll fell so disproportionately on the elderly and those with serious co-morbidities. As a result, many COVID victims had limited life expectancies before the pandemic. As a result, for some time moving forward, there will be fewer people vulnerable to the traditional causes of death like heart disease, cancer and non-COVID pneumonia.
In statistics there is a concept known as the “reversion to the mean.” It suggests that dramatic variations from historic patterns are more likely to be followed by patterns that are more typical. Wolfram Mathworld explains it: “an extreme event is likely to be followed by a less extreme event.” If that proves to be the case in this instance, we could see a significantly lower fatality rate in the US over the next year or two.
Let’s hope and pray that is the case.
Note 1 – I have a slight quarrel with the CDC methodology for calculating what the normal number of fatalities should be. I believe that they have insufficiently adjusted the numbers for the projected normal growth in fatalities. That growth averaged 1.3% from 2014-2019 but the CDC formulation came up with a lower number. However, the difference is not statistically significant. While the CDC shows average excess fatalities during 2020 and 2021 just over 20%, my calculation came a little under 19%.