The Misleading COVID Numbers – An Introduction

To really understand the nature of the COVID-19 “pandemic,” and to properly evaluate the official response, we need to gain some familiarity with the terms and figures related to infectious disease mortality. For my purposes here, I will momentarily shelve my doubts about the reality of the SARS-CoV-2 virus as a novel, lethal virus, and my lack of confidence in the RT-PCR test that has been the primary means of diagnosis.

This task can get confusing, given that there are a number of subtly different terms that tend to be used interchangeably. For example, the word “mortality” refers to the susceptibility of a group to death, or death by a certain cause; whereas “lethality” refers to the ability of something (e.g., an infectious agent) to cause death. So, in the case of an infectious disease, lethality is the flip side of mortality. Lethality also has a specific, technical meaning when it comes to an epidemic – it is the ratio of those who are exposed (i.e., placed in contact with the infectious agent) who will go on to die.

Generally, when we want to know the seriousness of an epidemic (and a “pandemic” is a global or near-global epidemic), we are going to look at the total number of people believed to have died as a result of the associated disease, and then we are going to divide by the relevant subset – known or estimated – of the population.  There are really three major ways of doing this, and thus three terms we need to be comfortable with:

• Case Fatality Rate (CFR)
  • the total number of deaths (in the target population) due to COVID-19 (or, more generally, the infectious agent), divided by the number of cases 
  • cases generally include symptomatic patients who sought treatment through the medical system, plus those treated following a positive screening test
    • complicated by the incredibly general (e.g., cough, fever, shortness-of-breath, itchy skin) symptoms associated with COVID-19; doctors have great leeway to diagnose COVID-19 simply based on symptoms, in the absence of testing or even in the face of multiple negative tests
  • inevitably, there will be some highly symptomatic patients that don’t seek treatment, and fight off the disease on their own
    • of course, these cases won’t be counted as such
  • likely, there will also be some patients for which a COVID-19 diagnosis is missed
  • due to the uncertainty, CFR will generally be an estimate, and may be given as a likely range instead of a single value
  • “confirmed” cases are usually those cases for which there has been a positive test
    • however, this can get murky when a person is tested multiple times and the results are inconsistent; moreover, doctors are sometimes given leeway to “confirm” a case even though all or most tests have gotten negative results (in a later post, I will give proof that, at least in one public hospital, patients are being recorded as “COVID confirmed” when all COVID tests have come back negative)
• Infection Fatality Rate (IFR)
  • the total number of deaths (in the target population) due to COVID-19, divided by the number of people infected (i.e., exposed to the point where there’s some kind of immune response)
  • A.K.A. lethality
  • this number is really hard to nail down with any certainty, for a number of reasons:
    • many people will be exposed to COVID-19 and never know it, that is, they won’t suffer any ill effects; others will suffer only mild effects
      • the vast majority of such people won’t seek treatment, and authorities will probably never know these particular people were exposed 
        • the exception is if they are given a serology test following the exposure and found to have the sorts of antibodies that particular test is looking for, known or believed to be indicative of a previous exposure
    • the immune system is very complex and multi-faceted; it is quite possible that a sub-population will have been exposed to COVID-19, but have fought it off using different components of the immune system than are being tested for
  • as a result of significant unavoidable uncertainty, IFR is almost given as a likely range
• Mortality Rate
  • the total number of deaths (in the target population) due to COVID-19, divided by the total population in number of millions, hundred-thousands, or thousands
  • gives the number per each million (or other unit) of the entire population (or sub-population) being studied that perished from COVID-19
  • this is the measure I was using in my previous post (“The COVID Zone”), to compare the toll of COVID-19 in the U.S. versus the non-lockdown countries (Sweden, Japan, Belarus)
  • Since the population size is fairly static compared to the short lifecycle of an infectious disease like COVID-19, this rate will generally only climb over time
    • the exception being when a mistake, e.g. double counting of deaths, is discovered and the mortality rate is adjusted downward accordingly

CFR and IFR are often given as percentages. A raw value can be converted to a percentage by multiplying by 100. Thus, a CFR of 0.05 becomes 5 percent.

The Diamond Princess (DP) cruise ship outbreak provides a decent illustration of the differences between CFR, IFR and mortality rate. There were 3,711 people on board (including crew; there were 2,666 passengers), 712 cases (381 of whom were symptomatic), and 14 COVID-19 deaths (all of whom were passengers). Note that I have not included the number of infected people – this is because this figure is always an estimate. The DP outbreak is unusual in that it was contained to one ship and there was lots of testing done, so that the numbers theoretically should be more accurate than in a less controlled environment, such as a town or city. Yet, the high level of testing that was done means that a raw CFR will be lower than in a normal situation, because a lot of minor or asymptomatic infections that wouldn’t ordinarily be counted as cases, will be counted as such (thus puffing up the denominator of the calculation). In fact, the raw Diamond Princess CFR may be a better approximation of the IFR than a true CFR. However, this IFR is likely to be somewhat inflated because of gaps in testing and an unknown percentage of healthy individuals fighting off the virus without leaving a trace of it in their systems (thus depressing the denominator). Of course, in any event diagnostic tests (and especially the RT-PCR variety used for COVID-19) are subject to errors, so just because 712 passengers and crew tested positive doesn’t mean that’s how many had measurable levels of SARS-CoV-2 in their bodies; the real number is likely to be somewhat lower or higher than 712, depending on whether there were more false negatives than false positives or vice versa. Anyhow, for the purposes of showing the differences between the three death measures, I’ll count only the symptomatic cases in my CFR calculation, and I’ll assume testing caught all of the infections. Crunching the numbers, we get a CFR of 14/381 = 0.037 (3.7 %); IFR comes to 14/712 = 0.020 (2.0 %); and mortality rate is 14/3,711 = 3.8 per thousand people on the ship (5.3 per thousand for passengers). Before leaving this topic, I should note that there were a couple of skewing factors that likely exaggerated the case, infection and death numbers: a very old population of passengers (median age 69), and the cramming together in a tight space one gets on a ship. Moreover, a number of sources have claimed that the Diamond Princess was outfitted with a new 5G cellular network – intense exposure to a new kind of electro-smog tends to provoke distress in sensitive individuals, often leading to serious respiratory or other ailments. Nevertheless, only 301 symptomatic cases developed, and only 14 died.

Getting back to generalities… one thing to bear in mind with all these calculations is that their accuracy (and thus their worth) is dependent on the quality of the inputs.  That is, if the total number of deaths is clouded by significant uncertainly as to how many of them were really due to COVID-19 (versus a serious preexisting condition), or the total number of cases is skewed to an unknown degree by over-diagnosis, inaccurate lab tests, or limiting testing to only the sickest patients, then it really doesn’t matter how accurate the arithmetic is – the result is going to be of little or no use.

Indeed, there is plenty of evidence that the official COVID-19 death numbers are sketchy at best.  One clue is that the numbers vary so much by city, country and region.  Is it really likely that outbreaks in similarly populated areas will have such markedly different outcomes as we’ve seen?  For example, the carnage in New York City (over 19,000 “confirmed” and over 4,500 “probable” deaths at the time of writing) versus the minimal toll in Tokyo (less than 400 at the time of writing).  Then there’s teeming Nigeria with a total of barely 1,100 deaths in a country of 201 million people, versus Brazil with over 140,000 deaths in a nation of 211 million.  And Venezuela (mortality of 22 deaths per million pop.) has hardly been touched, whereas next door neighbor Colombia (510 deaths per million) has been hammered.  Some sources say that the Far East has been afflicted with a milder version of SARS-CoV-2 than has the “west,” which would explain the almost universally lower numbers in that part of the world (e.g. S. Korea with mortality of 8 deaths per million).  However, there are a number of European countries with low numbers, such as Germany, Norway, Finland, Denmark, and a host of eastern European countries. 

Other variations arise because of cultural factors (e.g., extended families living together, customs related to physical contact, etc.), demographics (e.g., Japan, Italy and Sweden) have very elderly populations, which should skew death numbers higher than for other nations) and differences in treatment.  Contrary to the mainstream narrative, a number of physicians, clinics, hospitals and hospital chains have found that hydroxychloroquine (HCQ), especially when combined with zinc supplementation, is an effective treatment for COVID-19. The American Association of Physicians and Surgeons (AAPS) has even sued the federal government for greater access to HCQ. (I’ll go into the HCQ controversy in detail in a later post.)

Probably the most commonly broadcast of our three measures is CFR.  Certainly, people are interested in this – if I get sick, how likely am I to die?  The nightmarish CFRs for Italy – often over 10 percent – were instrumental in creating a climate of terror surrounding COVID-19, and largely (along with ludicrously inflated models coming out of the UK’s Imperial College) responsible for providing the rationale for the draconian isolation policies that have become commonplace across the globe.  Why were the Italian CFRs so high?  There were certainly a number of reasons – a strained healthcare system, horrific air pollution, a large demographic of vulnerable elderly – but the primary reason was simply that generally, what was being reported was “confirmed” cases; since it was national policy to test only the sickest, hospitalized patients, a fairly high percentage of these cases would result in death.  Moreover, as has become standard across the globe, Italian authorities were very “generous” in attributing deaths “with” COVID-19 as deaths “by” COVID-19.  Making a bad situation worse the media-and-political-class-induced-panic caused a flight of foreign nursing home workers out of Italy (to beat border closures and other restrictions), leaving many older Italians without proper care. No doubt this flight of caregivers resulted in fatalities, which fed back into the oh-my-gosh-COVID’s-killing-everyone narrative to induce more panic and justify draconian responses around the globe. Too bad the mainstream press couldn’t be bothered to share this vital context.

Quite a few studies have estimated the lethality (IFR) of COVID-19, and the picture that is emerging is far less dire than that painted by political authorities and mainstream media.  For example, the diligent, astute researchers at Swiss Policy Research have taken a close look at the relevant studies and concluded that the IFR is about 0.1%, meaning that about one-in-one-thousand of those infected will perish.  Of particular interest is a study out of the University of Zurich which found that the standard serological tests that measure antibodies in the blood will only detect about one fifth of coronavirus infections.  Applied to the CDC’s cautiously high “best estimate” of 0.65% (dubiously raised from 0.26% in July), that would yield an IFR of about 0.13%.  If that figure scares the pants off of you, perhaps you should just curl up in the fetal position and never leave your bed.  Moreover, all these studies that are computing lethality are using the official mortality statistics, which are vastly inflated, as I will show.  It follows, then, that COVID-19 lethality is correspondingly lower.

[With the first two installments of my COVID-19 series, I think I’ve established, at least, that this “unprecedented” pandemic is in no way an existential threat and certainly doesn’t justify gutting the world economy or deleting basic human rights. With this task out of the way, in my next post I will take a closer look at where the “proof” of a killer, novel coronavirus came from.]