STUMP » Articles » Yes, COVID was a top cause of death for middle-aged people in 2021 » 6 July 2022, 21:54

Where Stu & MP spout off about everything.

Yes, COVID was a top cause of death for middle-aged people in 2021  


6 July 2022, 21:54

Here is the ars technica article pointed out to me by various people:

COVID was the leading cause of death in Americans ages 45-54 in 2021

COVID-19 was the third leading cause of death in Americans between March 2020 and October 2021, accounting for one in every eight deaths.

In that time frame, COVID-19 ranked in the top five causes of death for every age group of people older than 15 years. Between January and October 2021, the pandemic disease was the leading cause of death among people 45 to 54 years old.

That’s all according to a study of national death certificate data published Tuesday in JAMA Internal Medicine by researchers at the National Institutes of Health.

The study found COVID-19 caused roughly 700,000 deaths between March 2020 and October 2021. The pandemic disease trailed only heart disease and cancer, which caused roughly 2.15 million collectively in that time frame. The fourth and fifth deadliest afflictions in the US were accidental deaths—including car crashes, overdoses, and alcohol-related deaths—and stroke, which collectively caused around 624,000 deaths during that period.

Oh, wait. They only covered through October 2021.

In any case, thanks for catching up with me from two months ago

A reminder from an earlier post

This was my post from May 13, in case you had forgotten: Top Causes of Death for 2021: Heart Disease, Cancer, and COVID.

Here’s the ranking table:

To be sure, I had a bunch of deaths yet to be categorized, and I’m hoping to have almost completed data either today or next week. Yay.

I understand the JAMA folks take time to get stuff published, and I can just sit around, pull my data, and write to my blog. When people find errors, I can just edit and re-post. Etc. I’m not an academic, or an academic journal, so I’m not particularly concerned about waiting to get everything perfect before publishing. I want good enough.

Checking other claims from the article

However, I am skeptical of some of the claims being made in the article, because I’ve done some ranking tables and visualizations in my time, and I don’t think the results they think they’re crowing over is much more than that they restricted their results to a specific time period.

Back to the ars technica coverage:

The authors, led by Meredith Shiels, an expert in cancer epidemiology and genetics at the National Cancer Institute, broke up the time frame into two sections: the start of the pandemic in March 2020 to December 2020, and January 2021 to October 2021, the last month for which there was complete data. This revealed age-specific trends, likely driven partly by uptake of vaccines and other mitigation efforts.

In the 2020 period, COVID-19 was the second leading cause of death in people age 85 and over, but, amid high vaccine uptake in this age group, it fell to the third leading cause of death from January to October 2021.

So, that’s ars technica. Let’s see what the article itself says:

The increased ranking of COVID-19 as a leading cause of death in some age groups is consistent with a downward age shift in the distribution of COVID-19 deaths in the US in 2021 compared with 2020 perhaps driven by higher COVID-19 vaccination rates in 2021 in the oldest age groups.

The pandemic also has had indirect effects on other causes of death in the US. From 2019 to 2020, death rates increased for heart disease, accidents, stroke, Alzheimer disease, and diabetes. Potential explanations are fear of accessing health care or misattribution of COVID-19 deaths to other causes. Accidental deaths (including drug overdoses and unintentional alcohol poisoning), assault, and suicide remain major causes of death in the US, particularly in younger age groups; the pandemic may have contributed to some of these deaths.

I added the emphasis above. They’re just speculating.

Now, here’s this guy:

Well, dude, if that’s your claim, about “red states”, then I’m going to look at COVID deaths for age 35-44 and 45-54 by state, okay? And then map those rates.

Hell, to be nice, I can start the count from April 2021, as that’s when I (in the age 45-54 bin) was eligible for the vaccine, so maybe we can say it wasn’t fair to count any deaths from the Jan – Mar 2021 wave of COVID.

Order of magnitude counts

But first, I am going to point out a big item in COVID for age 35-44 deaths first.

In my old ranking table (and yes, I did see the WONDER database was updated today, and yes, I’m doing a quick-and-dirty data pull), COVID was the number two cause of death, and now that I’ve got essentially a full year of death, I’m pulling the top 15:

33,468 for the number one cause: accidents (which includes drug overdoses and motor vehicle accidents, the two largest categories)

15,972 for the number two cause: COVID

Here is the same sort of table for age 45-54:

Yeah, COVID is still the top cause with this data update, which is what I had expected. However, note that COVID, heart disease, and cancer are all neck-and-neck for number of deaths. This is not really much of a difference in order of magnitude.

State comparison – tile grid maps to see a pattern

As per usual, I’ll be using my favorite tool for visually seeing if there is any pattern to the COVID mortality for these age groups, by state.

Recall, given the claim that there is some sort of connection, whether it’s due to political alignment, or vaccination in general (and I’ll come back to that), I need to exclude COVID deaths before they could reasonably have gotten vaccinated. I do need to have pretty completed data, etc. I am going to take April 2021 – December 2021 deaths so that I’m sure I’ve got the COVID deaths complete and that I’m excluding the winter 2020-2021 deaths. Basically, I’m grabbing summer through fall COVID deaths, primarily, as spring 2021 had almost nothing going on.

Age 35-44:

Age 45-54:

And for those who don’t have a good time with maps, I’ll give you the “top ten” (that is with the highest COVID death rates for these groups) and you can go to my substack to get the spreadsheet to get the full list. Yes, there were too few COVID deaths in Vermont for age 35-44 for me to get data — fewer than 10 deaths in a data pull, and the result gets “suppressed”. That is, I can’t get the info if there is so few deaths that I might be able to start identifying individuals.

Now, to be sure, I am no political expert in alignment in every state. My understanding is that most places are kind of “purple”, but there are one-party places, like California and New York. (I am exaggerating for effect, to be sure.)

One aspect of COVID deaths, though, we haven’t heard about much lately, especially with regards to younger adults, is obesity and diabetes.

And, I’m not at all sorry, but what some of this map is reminding me of, but not all of it, is where Americans are known to be really obese.

Yes, I know that doesn’t get one booked to the TV shows or writing columns for The Atlantic like just pointing out that Americans tend to be pretty fat compared to people in many other countries, and that there is a connection between obesity and higher mortality with COVID, it seems. Nobody wants to hear that (or the high drug ODs, which preceded the pandemic and will likely continue after the pandemic).

It used to be that obesity would kill you in a slower manner, via diabetes, heart disease, or other problems such as an overloaded liver. But a more direct, and rapid, pathway to death via an infectious disease was not what we were expecting at all.

Related Posts
Video: U.S. Mortality Through the Pandemic, All Causes and by Age Group, Provisional through 2022
Mortality with Meep: Excess Deaths And Coronavirus
Childhood Mortality Trends, 1999-2021 (provisional), Ages 1-17 -- Good News for Young Kids, Not for Teens