COVID-19 Mortality Counting and Numerical Misinformation

By Russell Payne

The COVID-19 pandemic brought society to its knees earlier this year with daily deaths surpassing 2,000 a day in just the United States. Famously however, some have argued that we cannot let the cure be worse than the disease, which raises the logical question: just how bad is it? While official death tolls in the United states recently topped 240,000 as of the writing of this paper there are those who claim that this statistic is false and overblown. Theories range from broad conspiracy against freedom to simple incompetence at the coroner’s office but they all stand opposed to the notion that this pandemic really is as deadly as the media, academia, and the government have reported it to be. The following discussion will address the arguments questioning the death toll in America as well as those reaffirming the seriousness of the pandemic. Specifically, the focus will fall on the statistical and numerical arguments on both sides and, which one uses those numbers in the fairest, clearest, and most honest way.

The first of the arguments questioning the seriousness of the pandemic comes from the depths of the internet and claims that only 6% of the COVID deaths are legitimate (Zimmerman). This specific statistic comes from Robert Zimmerman’s blog. Zimmerman it should be noted is a science fiction writer. He cites One America News, a far-right news channels well known for peddling conspiracy theories and misinformation. They claim that this statistic comes from the CDC’s own website and the co-morbidities page, which they do not actually show in the clip provided (Zimmerman). What becomes clear on the co-morbidities page is that people do not actually die from the coronavirus, they die from symptoms of the coronavirus. To clarify a co-morbidity is when a disease or condition affects a patient at the same time as another disease or condition. While co-morbidities do include people with pre-existing conditions like diabetes, heart disease, cancer, etc. (who still may not have died without having contracted COVID-19) it also includes people who contract COVID-19 and develop pneumonia while sick and then die with both pneumonia and COVID-19. So, while some co-morbidities may be more complex situations than other in terms of attributing what really caused the person to die, excluding all co-morbidities is a deliberately misleading and inaccurate way to read this statistic. It is like saying that they died because their heart stopped not from the coronavirus. The fact is, everyone dies because their heart stops, and if COVID puts your pulmonary and circulatory system under stress leading to a heart attack is that not dying form COVID? Sure, it was not purely COVID that killed you, but it played an integral role and without having COVID you would not have died. So, this 6% statistic is actually arguing that we should only count people who die with COVID, but without a clear co-morbitiy, as dying from the Coronavirus. Thus, it should be clear that this is misleading at best, and really more akin to misinformation rooted in a total ignorance of medicine or how cause of death is attributed.

The second argument we commonly see questioning the coronavirus death total takes a similar, yet less malfeasant form in claiming that “nearly a quarter of people Colorado said died from Coronavirus don’t have COVID-19 on their death certificate” (Ingold). This source goes on to quote Gov. Jared Polis as saying, “what the people of Colorado want to know is not who died with COVID-19, but who died of COVID-19.” This article also provides absolute values stating that 878 of the 1,150 COVID-19 deaths were “due” to coronavirus while the others simply died with coronavirus (Ingold). This argument is much more credible than the previous one however it still engages in the same confusion over co-morbidity and the article does not really do any work to define the difference between dying from the coronavirus and with the coronavirus. Without this information we really cannot break down this statistic and it makes it hard to say that it is credible or not. With this being said the inclusion of the absolute numbers as well as the relative percentage is to the authors credit in that it lets the reader know exactly how many deaths they are really debating. Thus, in examining this argument we see that its numerical argument is not necessarily false, but it is perhaps misleading in a similar way to the Zimmerman article. For example, a patient may be hospitalized because of the coronavirus but die because of diabetic shock, which was brought on in part by the coronavirus. This person would not have COVID-19 on their death certificate and according to this article would have died with instead of from COVID-19. However, are we supposed to ignore a life truncated by the coronavirus just because the immediate cause of death was not strictly COVID-19? Besides the sloppiness of terminology, the underlying argument looks to make the death of the most vulnerable members of society worth less than someone who had COVID-19 on their death certificate. Thus, while we cannot say that the numerical information is incorrect per se but they do present the argument, which most people might find rather insidious, in a way that is far more appealing than a fuller picture of the situation would be.

The third argument claiming that deaths from the Coronavirus are exaggerated makes the case that the deaths form the coronavirus are unreliable because determining cause of death was already unreliable. The author here contests that “between 20 and 30 percent of death certificates nationwide were wrong before COVID-19” (Priest). Priest the goes on to attribute this to the fact that the requirements to become a coroner varies widely across states. He cites as an example that “a Vermont study found 51% of death certificates had major errors” (Priest). This data seems fairly reliable and well sourced and points the issue of overworked and hurried coroners doing rushed work in uneven conditions (Priest). Generally, this seems to be the most credible argument questioning the COVID death count and to its credit they are using data that predates the pandemic. So, it can be understood that these studies are free from the new politicization around death certificates. Of all the arguments questioning the death count this one represents the most credible in that its categories are clearly defined, and the numerical data has a specific set of objects, death certificates, that it references. It also focuses on the strain on the system itself and points to specific reasons for the errors that are then represented by the statistics. However, in doing so it also latently makes the claim that any numerical data in this situation is not to be trusted which leaves us with a choice: either we use the best data we can get knowing that it is imperfect or we just use no data at all. It is also unclear if the CDC takes their death counts based off of death certificates or some more uniform method, which would be important to understanding the issue and which the article does not touch on. Although this article is not purposefully misleading in the same way as some of the arguments we have examined it does have major logical and argumentative flaws that make it questionable at best.

Generally then, most of the cases that are being made in order to question the coronavirus death toll hinge on some sort of misinformation or ambiguity within the categories. They all rely on relative presentation of their data, i.e. 6%, 1 in 4, or 20–30%, and only sometimes are able to back these percentages up with absolute values. Behind the numbers also lies a level of ambiguity in the material itself especially regarding the articles making the case for not counting co-morbidities. These articles, from Zimmerman and Ingold respectively, largely rest their cases on the assertion that co-morbidities should not count as deaths from COVID-19. This argument however is questionable at best and is also not what you would be led to understand from the headline. Indeed, the line between dying from COVID-19 and with COVID-19 is blurred especially when we begin to consider that many of the people dying with COVID-19 could have expected to live longer than they did not contract COVID. It seems like people’s deaths are being lessened simply because of a pre-existing condition, which was also one of the big concerns at the beginning of the pandemic. The final article makes the best case for questioning the CDC’s death toll although it is unclear whether or not that toll is coming from the coroner’s reports or from hospitals. Thus, even though Priest’s article is well researched it is not clear if it is really applicable to this particular situation. On the other side, news outlets and arguments supporting the current death toll and encouraging people to take the pandemic more seriously tend to focus more on absolute values as well as ways that we can double check the CDC’s count at a glance.

The first argument supporting the CDC’s figure looks directly at the 6% statistic that was presented by OAN and Zimmerman. This article examines the argument, in much the same way as above, and explains that “the truth is that in 6 percent of deaths, COVID-19 was the only cause of death listed and in the remaining 94 percent, people had at least one additional factor contributing to their deaths” (Patel). This is the same conclusion presented above and really goes back to why the coronavirus was reason for alarm to begin with, because it presents a serious danger for people with pre-existing conditions. The article then goes on to state that “we just reached 1 million deaths globally” and are “still seeing approximately 1,000 deaths a day” (Patel). Thus, not only does this article do a nice job of explaining why the 6% statistic is misleading it then goes on to provide the best data available on the situation. They also go on to explain that these people dying with coronavirus are dying in the face of “medical technology” meaning “increasing numbers of Americans are able to live long lives with pre-existing conditions” (Patel) but are instead dying of COVID. On the whole this article does a concise and easily understandable job of refuting the 6% statistic while also providing the absolute data on which that statistic was based. Although they do not necessarily justify the official WHO or CDC numbers beyond their regular legitimacy they do provide contextual evidence, that being the advanced medical technology we have to treat COVID and other diseases, as to why this pandemic should be taken very seriously. The clear use of the data available will becomes a reoccurring theme throughout the articles supporting the CDC’s count.

The second argument encouraging people to take the pandemic more seriously, also from NBC, makes the case that the current surge in COVID cases will lead to an even larger surge in the winter. They first cite the stat that the US recently “topped 1,000” deaths per day (Edwards) and that there “isn’t much we can do to prevent what we think is going to be about 2,200 deaths a day by Jan. 1” according to Dr. Vin Gupta a professor of UW Medicine. The citation of medical experts goes a long way in crediting this numerical argument, which also benefits from the fact that it is an absolute value and hard to construe. They easily, for example, could have stating that there would be an over 100% increase in the number of deaths per day between now and January but instead opted to present the absolute values. Although the article paints a rather grim picture of the winter it also does a good job of backing up its claims with expert opinions and clear statistics. This makes their claims more credible and although their prediction may or may not come true, they have presented the argument for why they believe it will very clearly. Here again we see the trend that generally the arguments supporting the statistics from the CDC are better sourced and clearer in their argumentation.

The third argument, coming from the Scientific American, shows its data in terms of absolute numbers and does it in a way that makes it very clear exactly how coronavirus deaths stack up against other leading causes of deaths. Zhou’s graph shows that it COVID-19, after seeing a massive surge earlier this year has settled (sort of) underneath heart disease and cancer in terms of deaths per week in the United States. It also clearly shows that COVID-19 is much more deadly that the flu. This graph is a nice visualization and presents the data in a clear and easily understood format, which should be equally accessible to readers of all education levels. There are also a number of other graphs that present useful information like historic deaths per 100,000 people along a timeline going back over 100 years (Zhou). This graph is useful because it shows that there has been a big surge in just total deaths making it hard to contest that the death count is fake, or that people are counting deaths from other causes as coronavirus fatalities. Each of the graphs also has sources cited for the information that they are charting so it is possible to go and confirm that these graphs are accurately portraying the information in question. Ultimately, these graphs implement numerical data in a clear, honest, and transparent way that leaves little to be desired. One thing that would be useful is if their graphs were updated like they are on the NYT coronavirus tracker so that we can see how they evolved past the point where this article was published. To Zhou’s credit the text of the article also does work to try to depoliticize the issue by praising an act signed by Richard Nixon that still stands as a progressive piece of public health legislation. In any case, this article does a nice job of presenting the data in an honest and clear way and does not really fall into any of the misinformation pitfalls that are common in statistical presentations.

The final argument and clearest argument backing up the CDC’s statistics on total deaths from COVID-19 come from STAT News. The article states simply that “nearly 300,000 more people died in the United States from late January to early October this year” (Joeseph). As mentioned in the article this statistic suggests that the CDC’s numbers are perhaps not totally inclusive, although it should be noted that the 300,000 excess deaths also includes all deaths beyond what we would expect based on extrapolating the data “from 2015 to 2019” (Joeseph). The article also does a nice job of breaking down the increased death rates among demographic groups as we see minority communities are hit hardest by the virus. Although 300,000 is probably an overcount for strict COVId-19 deaths the article highlights an important point: over a quarter of a million more people died this year than we expected, if it was not COVID-19 then what was it? To its credit, the article also notes a limitation of the study, that the population is growing and so we might have gotten more deaths than we expected anyways (but not quite 300,000).

As for the arguments supporting the CDC’s current count and more specifically urging people to take the pandemic more seriously there is an abundance of well-presented data. The main flaw with them is that they engage in speculation, which certainly strays from the world of hard data but, at least in these examples they back up their speculation with expert opinions and the reasoning behind why they are making their grim predictions. Thus, although the side urging people to take the coronavirus death count, and the pandemic generally, more seriously does have some areas where the numbers provided are speculative they do a generally better job of presenting the facts of the matter than those who would downplay the death count. They mainly present absolute data and go based off of official reports, which may be flawed some of the time, but are still explicitly for the purpose of measuring the impact of the pandemic and they are the best we have.

This then brings us to another issue in the debate which is really a categorical argument. Put simply the two sides are debating about what qualifies as a death from COVID-19. The first set of arguments largely claim that we should only count the deaths of people without preexisting conditions and who had no comorbidity towards the official death count. The other side includes those people who had preexisting conditions and might have died from the coronavirus exacerbating something that they already had issues with. Should we count only those cases where the person was otherwise healthy, or should we count all lives that were cut shorter because of the virus? Another question is, do we count deaths from symptoms of the coronavirus and conditions brought on by it? On one side the answer seems to be no while the other side answers yes. If someone came down with the flu and then died from pneumonia, which they got because of the flu, does that count as a death from the flu? Ultimately, these are questions for each individual to decide but it is clear that the CDC counts these deaths and that, as many Americans have some sort of COVID risk factor, perhaps everyone should. Ultimately however, and to their credit, the CDC is fairly transparent with their data and this transparency has created this debate.

With all of this being said, while there we will probably never really know how many lives the coronavirus will take and how many it already has, we can and must do our best to accurately count and report the data. What is clear is that many more people have died this year than anticipated and, although the exact number may be up for debate, to claim that this is not due to the coronavirus is denial. Articles like the Zimmerman article, or the Colorado Sun article really do mislead people into believing that the CDC is lying to the public, which it is not. Specifically, the 6% claim really is a real and dangerous stretch. To claim that someone who died from respiratory failure, while they were in critical condition from COVID-19, did not die from COVID-19 is misleading to the public and something most people would not accept. This sort of strange categorization does not appear to happen on the other side at least to the same extent. Although importantly it does come back to the question of what we say counts as a COVID-19 fatality, which is clearly defined on the CDC’s website. For this reason, we should take arguments on both sides while understanding the facts as best we know them and calling out misinformation when we can. Any way you choose to classify deaths, they are still deaths and it is wrong to put others in danger by making light of another’s untimely death.

Works Cited

Ingold, John, and Jesse Paul. Nearly a Quarter of the People Colorado Said Died from Coronavirus Don’t Have COVID-19 on Their Death Certificate. 15 May 2020, coloradosun.com/2020/05/15/colorado-coronavirus-death-certificate/.

Zimmerman, Robert. More Evidence the Number of COVID-19 Deaths Is Greatly Exaggerated. 28 Oct. 2020, behindtheblack.com/behind-the-black/essays-and-commentaries/more-evidence-the-number-of-covid-19-deaths-is-greatly-exaggerated/.

The New York Times. Covid in the U.S.: Latest Map and Case Count. 3 Mar. 2020, www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html.

Patel, Dr. Kavita. The Truth about CDC’s Covid-19 Death Rate — and the Conspiracies Undermining It. 29 Sept. 2020, www.nbcnews.com/think/opinion/truth-about-cdc-s-covid-19-death-rate-conspiracies-undermining-ncna1241343.

Priest, Jessica. One in 3 Death Certificates Were Wrong before Coronavirus. It’s about to Get Even Worse. 11 May 2020, www.usatoday.com/story/news/investigations/2020/04/25/coronavirus-death-toll-hard-track-1-3-death-certificates-wrong/3020778001/.

Zhou, Youyou. COVID-19 Is Now the Third Leading Cause of Death in the U.S. Scientific American, Scientific American, 8 Oct. 2020, www.scientificamerican.com/article/covid-19-is-now-the-third-leading-cause-of-death-in-the-u-s1/.

Edwards, Erika. With Covid-19 Cases Surging, the U.S. Is in for a Dark and Deadly Winter. 8 Nov. 2020, www.nbcnews.com/health/health-news/covid-19-cases-surging-u-s-dark-deadly-winter-n1246740.

Joeseph, Andrew. Amid pandemic, U.S. has seen 300,000 ‘excess deaths,’ with highest rates among people of color. 20 Oct 2020, https://www.statnews.com/2020/10/20/cdc-data-excess-deaths-covid-19/