Excess Deaths

By Bruce Webber
Posted 2024-03-08 in excess deaths

---

Since 2020, there have been excess deaths in the United States (US) and other Western countries. Excess deaths are the difference between the number of deaths that occurred and the number of deaths that are expected.

At the start of the pandemic, most of these deaths were from COVID. In 2023, however, the majority of these deaths were not from COVID. In the US, there were:

• 283,425 excess deaths (9.2% of total deaths), and
• 74,995 deaths attributed to COVID.

Subtracting, we see there were:

• 208,430 non-COVID excess deaths in the US in 2023 (6.8% of total deaths).

I would like to know:

• What are the causes of these non-COVID excess deaths?
• Have you heard of this? If not, why not?

These were not just older people. There were many excess deaths in the 0 to 44 age range.

These statistics are from the Organisation for Economic Co-operation and Development (OECD). Below, I describe how I obtained them.

Visualizing the Data

You can see charts of excess deaths at Our World in Data. Unfortunately, Our World in Data does not categorize the deaths by age.

To get greater detail, I downloaded mortality data from the OECD and created a Google sheet. It displays the data for the US and several other countries.

To create the sheet, I downloaded the data from the OECD Data Explorer on 2024-02-25. I then flattened the data so there is one row per country per week, with a column for each value displayed in the charts. The resulting data is in the OECD_excess_deaths tab in the Google sheet.

If you want to validate what I did, you can download the data from the OECD in CSV format and compare it to the data in the Google sheet.

You can also see the script I used to flatten the data in this Bitbucket repository along with the file from the OECD and the flattened file.

Statistical Considerations

The OECD provides a Methodological Note describing how it collected and calculated the data. It calculates excess deaths by "comparing the number of deaths recorded for a week against the expected number of deaths over the same week... The expected number of deaths is based on the average number of deaths for the same week over recent years (in this case the previous five years, 2015-19)." Using 2015-2019 as the baseline allows the comparison of deaths before the pandemic to deaths during and after the pandemic.

The OECD warns that "Reporting of the number of All-cause and COVID-19 deaths particularly for the most recent weeks may be only partial and subject to significant revision. The calculated values for excess deaths for the most recent weeks are therefore also subject to significant revision." We see that in the Percent Excess Deaths by Age chart, where the values drop off in the second half of 2023.

The Methodological Note points out that population growth makes the calculated excess death percentages larger than they should be, because the baseline is from 2015 to 2019. Here are recent US population statistics:

The population has grown by 1.74% since the start of 2020. This accounts for a small part of the 9.2% excess deaths reported in the OECD data.

---

Since writing the above, I published a second article about excess deaths.