Day 78: COVID-19 Post Hoc Ergo Propter Hoc

 

Our current snapshot above. But New York appears to have cleared the apex:

The deaths attributed to COVID-19 in NY will continue to rise for the next few days even as new cases fall.

The post today is prompted by a comment by @ontheleftcoast that linked to a video of Montana physician Dr. Annie Bukacek discussing how COVID-19 death certificates are being manipulated. Briefly, Dr. Bukacek describes how there is a distinction between:  (1) people dying from some other cause while having detectable levels of the COVID-19 virus in their system, (2) people dying where COVID-19 exacerbated a condition from which the person would die fairly soon but maybe not as soon as they did, and (3) where COVID-19 is the cause of death to a normally healthy person without comorbidities (what I will call “true COVID-19 fatalities”). Think of number (3) as being the equivalent of the young people who died in 1918-1920 from the Spanish flu pandemic. How prevalent is that?

At some point we are going to know how many people were exposed to the virus and how many people were infected by the virus. That will form the denominator for post-event assessments on the lethality and severity of this disease. But Dr. Bukacek is highlighting that CDC guidance on how and when to list COVID-19 on death certificates makes cause of death presumptively COVID-19 if the person presents with any of the symptoms or has a positive test and subsequently dies. This means that the numerator for the infection fatality rate is most certainly going to be overstated. In other words, not only do the models project a level of death, but the procedure for assigning the official cause of death will tend to support the predictions whether or not it is factually so. Post hoc ergo propter hoc — “after this, therefore because of this” — we predicted people will die of COVID-19, therefore they must have died of COVID 19.

Why should we care if the cause of death and/or diagnosis is wrong? Three reasons: (1) a lot of important public policy decisions are being made on the assumption that this virus has a certain prevalence combined with lethality, (2) we can’t improve our epidemic modeling if the data is significantly wrong, and (3) mischaracterizing cause of death injects error into all other data regarding mortality. The third reason is real, but as Dr. Bukacek points out a substantial number of death certificates already convert mere educated guesses into solid statistical facts.

Let’s consider what we know and don’t know about the contagion and lethality of the COVID-19 virus. Here is a chart that was included in a New York Times piece last updated on February 28:

Left to right is contagion from lower to higher. Down to Up is lethality from lower to higher. Recent reports about aerosolization — that the virus can be transmitted through exhalation without riding on a droplet from a sneeze or cough — moves the salmon-colored box further to the right than where it was placed on February 28. But note that on the chart (a logarithmic presentation) the infection fatality rate estimated on February 28 was from 0.1% to more than 5%. But as of this date we still have no idea how many people have been infected, we only know the number of people who either were diagnosed based on presenting symptoms or who tested positive for the COVID-19 virus even though asymptomatic. The latter group is pretty small because, as a general rule, tests are not currently being administered to anyone who is asymptomatic.

The case fatality rate (CFR) (as opposed to the infection fatality rate shown on the graph above) in the US is currently greater than 3% because dividing the current death toll by the current case total (3.2%) does not reflect how many of the new cases will end in death. But the CFR calculation only reflects people diagnosed or testing positive, not those infected but not presenting symptoms or experiencing symptoms so mild that they do not seek medical care for the condition. And CFR is also being overstated because COVID-19 is being listed as the cause of death routinely now for cases that would have listed a different morbidity in the past.

So the truth is we still don’t know with any precision what the likelihood of infection is or the likelihood of severe illness or death in the event of infection. We do know some of the risk factors that lead to more severe cases and death. Foremost amongst these is being very old.

But what is “old”? We all have both a chronological age and a biological age. We all know our chronological age but few of us know our biological age. If we knew the biological age of true COVID-19 fatalities, what would that tell us about the disease?

[Note: Links to all my CoVID-19 posts can be found here.]

Published in General
This post was promoted to the Main Feed by a Ricochet Editor at the recommendation of Ricochet members. Like this post? Want to comment? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

There are 32 comments.

Become a member to join the conversation. Or sign in if you're already a member.
  1. Kozak Member
    Kozak
    @Kozak

    iWe (View Comment):
    Which means that we cannot be sure whether COVID or our own auto-immune reaction to it is the cause of death.

    This about the best example of a distinction without a difference I have ever seen.

    • #31
  2. MISTER BITCOIN Inactive
    MISTER BITCOIN
    @MISTERBITCOIN

    I thought the Spanish Flu was deadliest of them all.

    1. it killed young people, ages 20 – 50
    2. people would turn blue and die within 72 hours of infection
    3. the number of deaths could have been worse except the virus killed hosts faster than it could find new hosts

     

     

    • #32
Become a member to join the conversation. Or sign in if you're already a member.