May 4, 2022
Why is there an egregious double standard when looking at Covid-19-related deaths vs vaccine-related deaths?
A number of studies have looked at potential vaccine harms, including fatalities. Other studies have examined “breakthrough” Covid-19 deaths, which are defined as Covid-attributed deaths in fully vaccinated people. The standard of causality used in studying these deaths, including requiring a direct causal linkage that is “proven beyond doubt,” is an extremely high evidentiary standard. Conversely, the standard of causality for attributing deaths to Covid-19 more generally has been extremely loose. If we use the same high standard in looking at “Covid deaths” more generally we find strong support for the conclusion that “Covid deaths” have been significantly exaggerated
Authors of this essay: Tam Hunt, J.D., John McGowan, PhD
CDC director Rochelle Walensky stated in early 2022 that the large majority of vaccine “breakthrough” deaths from Covid-19 were from preexisting conditions (comorbidities) in these victims. Breakthrough deaths are when a fully vaccinated person dies from Covid-19. Most of these deaths were not, however, Walensky argued, from Covid-19 itself.
These were, accordingly, deaths “with” Covid-19 rather than “from” Covid-19, occurring in already quite sick people.
She stated:
[CDC published] a study of 1.2 million people who are vaccinated between December and October. And demonstrated that severe [Covid-19] disease occurred in about 0.015% of the people who were — received their primary series — and death in 0.0003% of those people. The overwhelming number of deaths, over 75%, occurred in people who had at least four comorbidities. So really, these are people who were unwell to begin with.
The study itself (Yek et al. 2022) concluded (p. 24): “COVID-19–associated outcomes occurred in a small proportion of persons (0.015%) who had completed primary vaccination, all of whom were aged ≥65 years, immunosuppressed, or had other underlying conditions.”
The “over 75%” figure quoted by Walensky is actually 78% in the study and it applies to “risk factors,” which is a similar but more general term than “comorbidities” (p. 23). The risk factors studied by Yek et al. include (see Table 1): “age ≥65 years, diabetes mellitus, immunosuppression, chronic kidney disease, chronic liver disease, chronic neurologic disease, chronic cardiac disease, or chronic pulmonary disease.”
The same logic stated by Walensky and by Yek et al., applies, however, for all “deaths involving Covid-19,” and not just those associated with breakthrough cases. But it’s even worse for Covid-19 deaths more generally because there are actually more comorbidities in this much larger group than in the breakthrough cases studied in Yek et al., 2022.
CDC’s own data show that on average there were 4.0 other causes of death (comorbidities) in US “deaths involving Covid-19” (this is how CDC describes what the media generally just call “Covid-19 deaths) (“For deaths with conditions or causes in addition to COVID-19, on average, there were 4.0 additional conditions or causes per death”; see Table 3).
The list of comborbidities (“conditions contributing to death where Covid-19 was listed as a cause of death”) in this CDC table is a little longer than the eight “risk factors” examined by Yek et al. 2022 but is quite similar.
So it’s not only 78% of deaths that had 4.0 additional comorbidities, as CDC found in the context of breakthrough Covid-19 deaths, but 100% of deaths being attributed to Covid-19 more generally have had an average of 4.0 additional comorbidities.
This is similar to the recent debate, which is a major step in the right direction, about hospitalization “with” or “from” Covid. Walensky herself, again, and joined by Dr. Fauci, have highlighted since late 2021 the need to distinguish hospitalizations figures in terms of incidental (“with”) vs. causally-related (“from”) Covid hospitalizations.
One article explained:
The CDC also said during the briefing that the surge [in pediatric Covid-19 hospitalizations] could be partially attributable to how COVID-19 hospitalizations in this age group are defined: a positive virus test within 14 days of hospitalization for any reason.
The severity of illness among children during the omicron wave seems lower than it was with the delta variant, said Seattle Children’s Hospital critical care chief Dr. John McGuire.
“Most of the COVID+ kids in the hospital are actually not here for COVID-19 disease,” McGuire said in an email. “They are here for other issues but happen to have tested positive.”
Almost all patients in the US are tested for Covid-19 upon admission to a hospital, regardless of symptoms and regardless of why they’re at the hospital. They are also mandatorily tested before surgeries and often every few day sduring their stays. This practice of screening asymptomatic patients will necessarily lead to a large majority of false positives and a lot of incidental Covid-19 diagnoses, a problem explored by The Atlantic’s David Zweig in this piece.
One study discussed by Zweig, Webb and Osburn 2021, found fully 87% of pediatric “Covid hospitalizations” were not related or were only minimally related to Covid-19. To be clear, these pediatricians found that only 13% of the “Covid hospitalizations” their hospital staff had tallied were in the hospital because of Covid-19, and the rest were best described as hospitalizations “with” Covid-19 or, more accurately, simply a positive test result that in many or most cases was simply a false positive test result. (Webb and Osburn do not make any statements about false positives in their paper, but the mathematics of hospital Covid-19 screening dictate that a large majority of positive test results will be false positives).
Harvard Med School professor Dr. Branch-Elliman explains the logic of high false positives from widespread screening for Covid-19 in this US News & World Report essay. She and her coauthors calculated that up to 71 out of 72 school Covid-19 screening test positives would be false positives at the 0.1% background prevalence they observed in mid-2021. In other words, 98.6% of test positives would be false positives.
This same logic applies to “deaths involving Covid-19” more generally, and again for the same reasons: massively over-inclusive definitions that rely in most cases only on a positive test result to label someone a “Covid hospitalization” or a “Covid death.”
We see a similar double standard in tallying vaccine-related deaths
A number of studies have come out looking at deaths attributed to the Covid-19 vaccines themselves, due to side effects from the jabs. However, almost every study looking at this issue has applied a very stringent set of criteria for making a determination that a given death was caused by a Covid-19 vaccine rather than other causes.
A recent study, Schneider et al. 2021, published in the International Journal of Legal Medicine, looked at 18 deaths in people who were recently vaccinated for Covid-19. They concluded:
In 13 deceased, the cause of death was attributed to preexisting diseases while postmortem investigations did not indicate a causal relationship to the vaccination. In one case after vaccination with Comirnaty, myocarditis was found to be the cause of death. A causal relationship to vaccination was considered possible, but could not be proven beyond doubt.
A similarly high standard was applied by FDA and CDC in another study. An article “debunking” claims of over 1,200 deaths associated with the first three months of the Pfizer mRNA vaccine focused (correctly) on the need to establish a “causal link” between the vaccines and the reported deaths.
FDA spokeswoman Alison Hunt (no relation) highlighted the very high bar for establishing causal linkage, however, between vaccines and alleged harms: “Reviews by FDA and CDC have determined that the vast majority of the deaths reported are not directly attributable to the vaccines,” she said, adding that reports of deaths after Covid-19 vaccination are “rare.”
Similarly, FDA’s regulatory analysis of Pfizer’s Comirnaty vaccine and deaths that occurred during the clinical trials. This summary report was released under court pressure after FDA had already approved the Comirnaty vaccine and had refused to share its process documents with the public. The public FDA report stated that in the blinded portion of the Pfizer trial there were 21 deaths in the vaccinated group and 17 deaths in the unvaccinated control group. FDA concluded, however, that “none of the deaths were considered related to vaccination.” And there was no “causal relationship to Comirnaty” (FDA report, top of p. 23).
A government panel in India used similar logic to argue that deaths associated with vaccines were not causally linked to the vaccines, but were instead due to preexisting conditions (comorbidities):
“Prima facie evidence doesn’t show any causal linkage between deaths and the vaccines given so far in India. Most of the deaths happened with individuals who have other comorbidities like heart diseases, brain disorders, kidney disorders, blood pressure, diabetes etc. It does not look like there is a spurt of occurrence of these events.”
The double standard is stark: “proven beyond doubt,” need for a “causal link,” “causal relationship,” and “directly attributable,” when examining possible vaccine-related deaths but a far far looser standard for Covid-19 deaths more generally, which I’ll look at in more detail in the next section.
Another example in a related field can be found in a June 2022 fact check regarding hospital-caused deaths from healthfeedback.org. The fact checkers discussed a study by surgeon Marty Makary and offered the following critique of the study: “it didn’t use dataset or analysis methods that could determine whether adverse events in patients who died [from medical error] were actually the cause of death.” They also state that: “the authors used a dataset from a group of people that aren’t representative of the general population. For instance, one of their dataset came from Medicare patients, a large majority of whom is over 65. This age group is at a higher risk of death than the general U.S. population and thus cannot be considered to be representative of the entire population. Therefore, this study cannot be used as reliable evidence to support the claim.”
Again, if we apply this same kind of causal standard of evidence to Covid-19 deaths, and appropriate demographic analysis, we would obtain a very different figure than the public statistics we are offered for Covid-19 deaths.
Figure 1 summarizes the various standards looked at in this essay.
What is the standard for reporting a Covid-19 death more generally?
Strangely, the World Health Organization’s Covid-19 death reporting guidelines, issued in April 2020 at the beginning of the pandemic, instructed death certifiers to ignore preexisting conditions — another term for comorbidities — in deciding whether a given death was caused by Covid-19 or not. The document states (p. 3):
A death due to COVID-19 may not be attributed to another disease (e.g. cancer) and should be counted independently of preexisting conditions that are suspected of triggering a severe course of COVID-19.
The same document instructed certifiers to list any death as a Covid-19 death when it was even suspected to involve Covid-19, or when Covid-19 was thought to have “contributed” to the death.
The US CDC death certification guidelines, issued in March 2020, were equally broad:
COVID-19 should be reported on the death certificate for all decedents where the disease caused or is assumed to have caused or contributed to death.
A more detailed guidance document from CDC, issued in April 2020, added:
If COVID–19 played a role in the death, this condition should be specified on the death certificate. In many cases, it is likely that it will be the UCOD [underlying cause of death], as it can lead to various life-threatening conditions, such as pneumonia and acute respiratory distress syndrome (ARDS).
So even an assumption that Covid-19 caused “or contributed” to death (in any manner, since no limitation is included for what qualifies as “contributed” or “played a role”) was enough to list Covid-19 as a cause of death on the death certificate.
These key guidance documents, issued shortly after WHO declared Covid-19 a pandemic on March 12, 2020, made it clear that, essentially, even a suspicion of Covid-19 being involved in a death was enough for it to be reported as a Covid-19 death. In the US, death certificates are collected and analyzed by the CDC and any certificate deemed to have a U07.1 international death certification code (CDC does this “coding” itself, based on written notes in the death certificates, a process that is largely automated by CDC’s computer system) was tallied as a “death involving Covid-19.”
How did we get to this double standard in death reporting?
So why did WHO and CDC apply such a loose standard for determining Covid-19 deaths for the duration of the pandemic and then apply such a remarkably more stringent standard for determining vaccine-related deaths and “breakthrough” deaths?
At this point it’s not clear at all how this happened. We don’t have to posit nefarious motives for this obvious and extreme double standard. It could readily be explained through “erring on the side of caution” in the beginning of the pandemic, which then built an unstoppable inertia of its own due to over-reporting and the appearance of massive numbers of deaths. And the later far more stringent standards for vaccine-related deaths and breakthrough deaths could be explained through an understandable desire to avoid inflaming vaccine doubters by rolling too many deaths into these categories.
Whatever the motivations were, and however noble those motivations may have been, what is clear at this point is that there has been an egregious double standard applied in these different categories. The general category of “deaths involving Covid-19” has been massively exaggerated in the US and around the world because of the very loose evidentiary standard for making this attribution. Conversely, vaccine-related deaths and breakthrough deaths may have been significantly under-estimated because of the very stringent standard applied in those contexts.
What do we do about this double standard?
How do we remedy this situation? A good first step would be to apply the same evidentiary standard in all three contexts examined above. In policy circles, this move is known as “harmonization” and it is a natural solution for a remedy moving forward. An example of harmonization in the pandemic context is the World Health Organization’s effort to harmonize excess deaths reporting across countries in an April 2022 report (“harmonized methods for excess mortality”).
Webb and Osburn 2021 suggest a three-part taxonomy for assessing whether a hospitalization is due to Covid-19 or not: 1) causally related to Covid-19; 2) minimally related; 3) not related. A similar approach could be used by each state in the U.S., or by the CDC in doing a nationwide approach, to reassess death certificates where Covid-19 was listed as a cause of death.
It is a common standard best practice to have scientific issues such as the proper assignment of causes of death evaluated by a transparent and blinded independent review. This is done to avoid both conscious and unconscious bias by parties involved. The apparent gross discrepancies between the standards for assigning the cause of death to the Covid-19 vaccines, or breakthrough deaths, versus Covid-19 itself is evidence that powerful biases are likely at work.
For these reasons, we recommend that a review of the assignment of Covid-19 as the cause of death, and also for respiratory illnesses like influenza and pneumonia prior to the pandemic, be performed by an independent group or groups of statisticians not employed by the CDC directly or indirectly, nor by other public health or governmental agencies involved in the CDC’s tallying of death and case figures. Multiple independent groups would be preferable for conducting such a review.
One possibility may be a group or groups of experienced actuaries drawn from the insurance industry. The independent review should be transparent and all data made publicly available with the possible exception of some personally identifying information. There should be timely opportunities for public comment and notice throughout the process.
Death certificates are data on the dead for whom privacy concerns are less of a concern than for the living. Given the seriousness of the pandemic and pandemic response involving the reported loss of hundreds of thousands of lives and expenditures of trillions of dollars in the United States alone, disclosing the actual names, dates of death and other information of decedents seems to us to outweigh privacy concerns for the deceased.
A key element of such an independent review should almost certainly a *blinded* review or series of reviews with different blinding methods in which the reviewers do not know the previous assignments of the underlying cause of death by the CDC, medical examiners, or others. This can be done by combining death certificate data from pre-pandemic years such as 2018 with data from the Covid-19 pandemic where Covid-19, pneumonia, and or influenza (the new “PIC” category used by CDC and others) is mentioned on the death certificates, certainly as a cause of death or contributing factor.
The original assignment of a death to Covid-19, influenza or other specifically identified bacterial or viral pathogens should be redacted (blacked out) along with the date of death or other dates that could show whether the death occurred before or after the Covid-19 pandemic began. The reviewers would be free, based on their best judgment, to assign the death to Covid-19, influenza, or other causes. The best comparison may be to compare pre-pandemic deaths where influenza is specifically detected to Covid-19 deaths that have a confirmatory PCR (with cycle threshold of 30 or below in order to ensure a strong positive) or other Covid-19 positive test result. Blinding would require using a neutral unrevealing term such as “pathogen detected with molecular test” that does not distinguish between SARS-COV-2, influenza, or other pathogens.
This is a *rough* outline of a potential study as we are not experts in conducting blinded studies. A more thorough blinded study design should be produced by the independent review group or groups and subjected to public notice and comment prior to performance of the study. A key metric would be the percentage of blinded death certificates where the unidentified blacked-out respiratory illness was rated as the underlying cause of death (UCOD) compared to the unblinded assignments where either COVID (pandemic) or “pneumonia and/or influenza” is listed as the underlying cause of death (pre-pandemic).
This kind of blinded ex post review of a sizable sample of US deaths would shed significant light on the degree to which assignment of Covid-19 as the cause of death was warranted or not, given reasonable criteria for such assignment.
(A later essay will look at more precise approaches that could be employed in establishing causal linkages between pathogens and deaths, such as Koch’s Postulates and Hill’s Criteria for biological causation.)
