How many deaths counted as Covid-19 deaths are actually from the virus?
A number of policy changes have made data quality a serious concern, possibly significantly over-stating (not under-stating) the severity of the pandemic in the United States
Part I: Data quality issues regarding U.S. deaths reporting guidelines
Two policy changes occurred in the beginning of the pandemic in the U.S. that seem to have led to serious data quality issues in a number of areas: 1) changing the longstanding guidelines for reporting cause of death, which led to reporting deaths from COVID-19 far more frequently than under the previous guidelines that had been in place for 17 years; 2) substantially reducing the accuracy standard for coronavirus tests during the nationwide emergency.
This essay looks at the first issue and my next essay will look at the second issue.
CDC guidelines for reporting cause of death were loosened significantly at the beginning of the pandemic
First, the quick summary of my analysis: I analogize the societal response in the US to COVID-19 to the body’s immune system over-reaction in the most serious cases, in which “cytokine storms” induce the immune system to destroy its own organs and thereby kill the patient. If my analysis is correct we may, with our government policy reactions to the virus, be causing the lion’s share of the harm that is being attributed to the virus. This is not to deny that there is a real illness and a real pandemic. But it does suggest that the severity of the pandemic has been significantly inflated.
The National Vital Statistics System (NVSS) is part of the CDC and is responsible, as the name suggests, for tracking things like mortality data. Steven Schwartz, the NVSS director, issued a number of reports in March and April of this year that significantly changed guidance on how COVID-19 deaths should be reported in the U.S. Specifically, the guidelines required listing COVID-19 in Part I of the death certificate as a “cause of death” even if COVID-19 was only suspected as being a contributing factor.
No positive test for the virus was required. And the virus was not required to be the primary cause of death — it was enough to have “contributed to death,” with or without any testing, to list it in Part I. The March 24 report states (emphasis in original):
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.
The March 24 report also stated that “the rules for coding and selection of the underlying cause of death are expected to result in COVID-19 being the underlying cause more often than not.”
And this report adds a key statement: “If the death certificate reports terms such as ‘probable COVID-19' or ‘likely COVID-19,’ these terms would be assigned the new ICD code. It is not likely that NCHS will follow up on these cases.” In other words, there will be no review or accountability by NCHS for reporting Covid-19 as the cause of death even when there is doubt about the actual cause of death.
As we’ll see in a later essay on financial incentives for over-reporting Covid-19 deaths, this lack of accountability likely played a very significant part in exaggeration of the pandemic. The adage “follow the money” is often true.
A follow up April report by CDC, “Guidance for Certifying Deaths Due to Coronavirus Disease 2019 (COVID–19),” contained more detailed guidance from NVSS on this new death certificate reporting standard. It specifies that if COVID-19 contributed to the death it is likely to be listed as the underlying cause of death (UCOD) in Part I of the death certificate:
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, as it can lead to various life-threatening conditions, such as pneumonia and acute respiratory distress syndrome (ARDS). In these cases, COVID–19 should be reported on the lowest line used in Part I with the other conditions to which it gave rise listed on the lines above it.
The World Health Organization followed suit also on April 20 with its own detailed guidance document, “INTERNATIONAL GUIDELINES FOR CERTIFICATION AND CLASSIFICATION (CODING) OF COVID-19 AS CAUSE OF DEATH.”
This WHO document also contains the important guidance (emphasis added): “COVID-19 should be recorded on the medical certificate of cause of death for ALL decedents where the disease caused, or is assumed to have caused, or contributed to death.”
WHO implemented a new death coding standard, U07.1, on January 31, 2020, in an emergency meeting, before WHO issued its own detailed guidelines on death reporting (see bottom of the web page). The new WHO standard actually went into effect on April 1.
Here is an example of the trend toward extreme liberality in listing COVID-19 as a cause of death, from a medical doctor writing for other medical professionals in terms of how to use the new WHO death reporting code:
I don’t need to do a laboratory test for measles or mumps; if I see a patient and determine either is present due to the constellation of signs and symptoms in the context of a breakout, I diagnose it, and it should be coded. During this time of pandemic, COVID-19 is overwhelmingly prevalent.
It should be clear how this attitude toward the virus being “overwhelmingly prevalent” is a circular argument when it leads to medical professionals listing COVID-19 even when no test was conducted, and when the symptom list for COVID-19 keeps growing steadily. It doesn’t take much imagination to see how almost any death can be linked under such a loose standard to COVID-19.
Listing COVID-19 as a cause of death in Part I of the death certificate is important because it’s almost always the case that only data from Part I is captured in national statistics and thus constitutes the single “COVID deaths” figure reported by many organizations.
In fact, CDC makes this reliance on the new code explicit in explaining its statistics, stating (emphasis in original):
COVID-19 deaths are identified using a new ICD–10 code. When COVID-19 is reported as a cause of death — or when it is listed as a “probable” or “presumed” cause — the death is coded as U07.1. This can include cases with or without laboratory confirmation.
Serious preexisting conditions (“comorbidities” or “premorbidities”) that are not thought to have been part of the specific causal chain that led to death are, under the CDC and WHO guidance, to be listed in Part II of the death certificate — and these data are generally not captured in national statistics, or at least not in the headline numbers.
The end result of this substantial change to death reporting guidelines is that any time COVID-19 is thought to have been either a cause of death, or a contributing factor in death, with or without testing, it will be tallied as a “COVID death” in national statistics tracked by various parties, including the COVID Tracking Project, Johns Hopkins, CDC itself, New York Times, and many others.
There is generally a single number reported for deaths in each state and each nation, with no distinctions between deaths more likely to be caused by comorbidities than the virus itself, as long as COVID-19 was listed in Part I of the death certificate as a cause of death OR a contributing factor to the death. And with the strong preference from authorities, stated above, to report COVID-19 as a cause of death even without any kind of testing, the potential for data quality issues is apparent.
This very loose approach to deaths tracking is summed up well by Dr. Deborah Birx, the White House Coronavirus Response Coordinator, in her April 2020 statement that “if someone dies with COVID-19 we are counting that” as a coronavirus death. This means that anyone dying “with” and not necessarily “from” (that is, causally connected to the virus rather than some other cause of death such as a serious comorbidity or preexisting condition), it is being tallied as a “COVID death.”
The 2003 CDC reporting guidelines were far more strict in terms of reporting contributing causes of death
The previous CDC reporting guidelines, in place for 17 years, required a higher standard for attributing a death to COVID-19. The previous standard, the 2003 CDC Medical Examiners’ and Coroners’ Handbook on Death Registration, directed medical examiners to report contributing causes of death in Part II (Handbook, p. 14): “Part II is for reporting all other significant diseases, conditions, or injuries that contributed to death but which did not result in the underlying cause of death given in Part I.” There was no exception to this guidance provided for any virus deaths reporting.
The new NVSS guidelines include similar language about putting other causes of death as contributing factors in Part II (rather than Part I), but (this is a very significant “but”) they also strongly suggest putting COVID-19 contributed to death entries in Part I — and only COVID-19 gets this exception.
(The key paragraph from the April NVSS guidelines is, with “other” referring to causes of death other than COVID-19 (emphasis added): “Other significant conditions that contributed to the death, but are not a part of the sequence in Part I, should be reported in Part II. Not all conditions present at the time of death have to be reported — only those conditions that actually contributed to death.”)
So this is the big change that led to a potentially significant over-reporting of COVID-19 deaths: a change from reporting contributing factors for deaths in Part II, under the 2003 guidelines, to reporting, specifically and only, COVID-19 as a contributing factor instead in Part I — which is the figure that data aggregation sites include in their running tallies.
The end result of this change was to shift what is very likely a large number of fatalities into the COVID-19 tally that would not have been in the tally under the previous standards. Or, to quote Dr. Birx again: “if someone dies with COVID-19 we are counting that.”
This potentially large over-inclusion of deaths as COVID-19 deaths makes calculating key pandemic statistics like the Case Fatality Rate and the Infection Fatality Rate very problematic, and it also risks significant over-reaction by policymakers if, in fact, the true death rate is significantly lower.
Comorbidities have been extremely common in hospitalizations and deaths counted as COVID-19 deaths
Various studies have found very high levels of comorbidities in COVID-19 deaths and hospitalizations, up to 94% of New York City’s hospitalized patients had at least one comorbidity, and 88% of patients had two or more comorbidities. A peer-reviewed study of Italy’s (the first hotspot for deaths after China) dead found similar, but even more pronounced, results, with 99% having at least one significant comorbidity and 74% having two or more.
The same study (Onder et al. 2020), which examined Italy’s high reported death rate and case-fatality rate, highlights the fact that Italy chose to define a COVID-19 related death as someone who died and had at some point tested positive for the virus with a PCR test, regardless of whether the patient died from the virus or something else. The study suggests: “Electing to define death from COVID-19 in this way may have resulted in an overestimation of the case-fatality rate.” The study also concludes: “The presence of these comorbidities might have increased the risk of mortality independent of COVID-19 infection.”
Italy’s health minister stated, based on a different study, that “only 12 percent of death certificates have shown a direct causality from coronavirus.” Direct causality means that the medical professional assessing the cause of death could say with some confidence it was the virus that caused the death, rather than “dying with” COVID-19. In other words, fully 88 percent of Italy’s dead were found to have died from something other than the virus since no “direct causality” was found from the virus to direct cause of death.
The U.S. has gone yet further than Italy in loosening its standards for the death count, as discussed above. Whereas Italy included in their death count any person who tested positive for a PCR test (considered to be the most accurate of the various tests available, but as we’ll see in Part 2 is actually not at all accurate as it is currently being used), official U.S. policy has been to include anyone testing positive for the virus under any test, or even suspected to have the virus, as COVID-19 related deaths.
This difference in reporting standards could be a large part of why the U.S. death count has been so much higher than other countries.
When we reflect on these reporting standards issues and observed high level of comorbidities in COVID-19 patients and deaths, we can see how there has probably been a substantial inflation in deaths reporting, despite an often-stated view that the opposite is probably true, though little evidence is usually provided for this often-stated view.
About half of all COVID-19 deaths in the U.S. have occurred in nursing homes or in hospital patients who got sick in nursing homes and were transferred to a hospital, and many of those deaths have, unsurprisingly, been associated with two or more comorbidities in each patient. Moreover, the median survival time of nursing home residents is only five months in normal times (pre-pandemic), which would have resulted in far more deaths, in the time since the pandemic began, from normal causes of death than are being reported now as COVID-19 deaths in nursing homes.
These dynamics suggest strongly that a large number of nursing home deaths being tallied as COVID-19 deaths would very likely have happened in the same timeframe without the virus being a factor at all.
A late August 2020 update from CDC quantified the number of comorbidities in U.S. deaths: there has been an average of 2.6 comorbidities in those counted as COVID-19 deaths: “For deaths with conditions or causes in addition to COVID-19, on average, there were 2.6 additional conditions or causes per death.” This does not mean (as many who read this misunderstood) that only 6% of reported deaths were from COVID-19. Rather, it means that there is significant doubt about the true cause of death in the vast majority of reported COVID-19 deaths. And, as discussed in detail above, official U.S. guidelines were to report COVID-19 as the underlying cause of death “more often than not” even when the virus was merely a contributing factor or suspected of being a contributing factor.
These considerations lead to the conclusion that the U.S. was essentially primed for a significant over-reaction to the pandemic, with medical professionals being urged by the relevant authorities to report any deaths remotely related to the virus, whether causally related or not, as COVID-19 related deaths, and the media seizing upon these early figures for headlines, further inflaming public sentiment and the vigilance of medical professionals to the prevalence of the virus, but without realizing the serious data quality issues associated with the reported mortality figures.
It seems to have been a vicious cycle of over-reaction feeding over-reaction — an interesting parallel to the cytokine storm phenomenon that is responsible for the worst-suffering cases of COVID-19, but at the societal level rather than the individual level.
Explaining excess deaths figures in the U.S.
We still have to address the crucial question: if it wasn’t the novel coronavirus that led to the significant number of excess deaths in 2020, what was (Figure 1)?
Excess deaths figures aggregated by CDC show that there is clearly a much higher death rate for 2020, but it is not the case that all of these excess deaths are caused by the virus.
A large number of excess deaths are due to causes other than COVID-19, as CDC’s data also show (Figure 2). These non-COVID-19 excess deaths total over 70,000 deaths — almost half of the total excess deaths reported in the U.S. so far (as of early August, 2020). Most of these excess deaths, and perhaps all, were arguably caused by the policy responses to the novel coronavirus — not the virus itself —and the media treatment of the virus, which has exacerbated underlying conditions, limiting hospital access, increasing loneliness and dementia, particularly in nursing home populations, and other causes, as described in the figure beow.
But it gets worse. A key disclaimer for this data is this statement from the CDC web page “methods” section: “Deaths with an underlying cause of death of COVID-19 are not included in these estimates of deaths due to other causes…. For the majority of deaths where COVID-19 is reported on the death certificate (approximately 95%), COVID-19 is selected as the underlying cause of death.”
So this CDC analysis of excess deaths does not take into account the fact that medical professionals were heavily influenced by NVSS guidelines and other factors to list COVID-19 as an underlying cause of death “more often than not” (as the March 24 NVSS report stated). Apparently these guidelines led to “more often than not” being 95% of the time. Accordingly, there is yet another significant data quality issue with CDC’s data on excess deaths from non-COVID-19 causes.
This data is more evidence that we have, as a society, engaged in a “metaphorical cytokine storm” set of over-reactions, possibly causing more harm to ourselves than the virus itself has caused directly.