COVID Deaths in the US: Who Pays the Highest Price?

This transcript has been edited for clarity. 
Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr F. Perry Wilson of the Yale School of Medicine.
From March of 2020 to May of 2023, 1.38 million more people died in the United States than would have been expected — 1.38 million lives that, had circumstances been different, would not have been lost. 
This is the legacy of the COVID pandemic. And I want to be clear that not all of these deaths are directly due to COVID; we’ll explore the deaths attributable to the virus in a minute. But the pandemic itself, the changes it wrought on society — the delayed cancer screenings, the missed outpatient appointments — all of these acted to change our fundamental understanding of the risk of living in this country.
And, of course, that 1.38 million number reflects the other side of this coin: a reduction in deaths from traffic accidents, for example, as fewer people were on the road. 
Perhaps we can look at those excess deaths and wonder what we could have changed to make a difference. Or perhaps we can look at that number and say, yes, this is tragic, but it’s the price we all paid for a global pandemic. Of course, that last statement would be easier to make if we all paid the same price. But as new research published in JAMA Network Open now shows, this is not the case.
Before we dig into the numbers, let’s talk about the concept of “excess deaths.” The idea is pretty simple: Researchers take historical death data and use it to model the expected number of deaths in the future. The models are fairly sophisticated, using monthly death data to capture seasonal changes, and accounting for demographic changes such as the aging population.
So, with decent accuracy, based on prior data, we can say that there should probably be 275,000 deaths from any case this month. And then you look at the observed deaths and the difference is your excess mortality, which of course could be negative, positive, or zero. 
It won’t surprise you to hear that during the COVID-19 pandemic, that number was almost always greater than 0. This graph shows you the excess death rate over time during the pandemic. You can see how the peaks of this metric align with the various COVID waves that crashed over the nation from 2020 to 2023.
You also see how closely the yellow line matches the blue line. The yellow line represents deaths due to COVID, so a large percentage of this excess is due to that infection. But not 100%.
Overall, this represents a 15% increase over expected death rates, meaning the number of people who died during this period was 1.15 times what we would have expected if conditions were normal. But what this study looks at is how this 15% was spread across society. And the findings are concerning.
Here are the expected and observed number of deaths across racial groups in the United States.
You can see that this 15% increase over baseline is not uniform, with those listed as American Indian or Alaskan Natives (a governmental designation) experiencing the greatest excess mortality relative to population size. Every group had higher-than-expected mortality, but every non-White population had higher excess mortality than the White population. 
I also need to point out that the modeling for expected deaths included race as a covariate. In other words, the model already takes into account that many non-White racial groups have higher mortality, so all of this effect is over and above that by now well-established difference.
The authors break this down by age group, but the pattern largely persists regardless of age. I’ll show you the youngest age group, zero to 24 years, since this is the only one where any group had a similar observed and expected mortality. But even in this group, it was only among White and Asian young people that no significant change in mortality during the pandemic was seen. All other races continued to show excess deaths.
We need to remember this when we hear people saying that COVID is a disease of the elderly or that it has no impact on young people. The truth is, it appears to have little impact on young White people — and because the majority of the population is White, that can blind us to some unsettling effects.
Of course, I can’t tell you why this is. Race is a social construct and provides minimal, if any, insight into genetics. As such, it would be odd for these findings to be due to some fact of biology. More likely, these findings are due to sociology, to the effects of societal stress on populations that may be more marginalized or have less access to care. 
We can take some comfort in the fact that mortality rates seem to have returned to normal by this point, which is to say the pre-pandemic disparities continue to exist. But we would do well to remember that whether it comes to paying the price for a pandemic or paying the price for any of our societal ills, those costs are rarely distributed evenly. 
F. Perry Wilson, MD, MSCE, is an associate professor of medicine and public health and director of Yale’s Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He posts at @fperrywilsonand his book, How Medicine Works and When It Doesn’t, is available now. 
 
Any views expressed above are the author’s own and do not necessarily reflect the views of WebMD or Medscape.