Context, category errors and usefulness

From Some Much-Needed Coronavirus Perspective. Its an editorial.

Four and a half months from discovery. Two months plus since its arrival in the US. A new disease that is little understood and for which there is little data. The models can be made to say anything. The models will say anything, even with the best of intentions. Garbage in, garbage out. We are months away from widely shared and standardized definitions of what constitutes death from Covid-19. We are probably realistically even more months away from having large data sets of comprehensive and reliable data from exposure to infection to recovery.

And part of this editorial is a category error. They are calling for perspective which is needed. But the best perspective is against the progression of earlier novel viruses such as H1N1 and the like.

Comparing new spreading disease to static known diseases and conditions is a category error. But it does provide a context to be kept in mind.

There are almost 140,000 active cases of COVID-19 in the United States. Nearly 20,000 new cases were reported on Sunday alone. The death toll in the U.S. is now close to 3,000 — with more than 2,000 of them occurring in just the past week.

National Institute of Allergy and Infectious Diseases director Dr. Anthony Fauci said on Sunday that the death toll could reach 200,000. Another model says 82,000 will likely die, with daily deaths peaking in mid-April at more than 2,000. By comparison, the last pandemic — the so-called swine flu — claimed 18,000 lives.

Appropriate policy responses are still guess work, even among the most clever, resilient, and competent countries. When the US, Norway, and Denmark head in one direction and Sweden 180 degrees in the other, it has to be acknowledged that something more than evil behavior or incompetence are at play.

With 6,000 dead and rapid transmission in different parts of the country, this is clearly a national emergency which needs considered responses when there is little established consensus on what the appropriate response ought to be. Of course, there is a discussion to be had whether the number is 6,000 or 3,000 or something higher or lower depending on definitions and how data is collected.

I&I provides context.

There are almost 140,000 active cases of COVID-19 in the United States. Nearly 20,000 new cases were reported on Sunday alone. The death toll in the U.S. is now close to 3,000 — with more than 2,000 of them occurring in just the past week.

National Institute of Allergy and Infectious Diseases director Dr. Anthony Fauci said on Sunday that the death toll could reach 200,000. Another model says 82,000 will likely die, with daily deaths peaking in mid-April at more than 2,000. By comparison, the last pandemic — the so-called swine flu — claimed 18,000 lives.

One pandemic to another is a reasonable comparison.

To a lesser extent, but still valid, is a comparison to all causes mortality.

On its own, 3,000 fatalities might seem like a tremendously large number. But that’s before you learn that an average of 7,700 people die in the U.S. every single day. Which means that over the past week, when the coronavirus took 2,000 lives, nearly 54,000 people died from other causes.

As a service to readers, here are recent annual deaths from other causes, many of which go largely unnoticed year by year, but most of which are preventable. (The data are compiled from the National Center for Health Statistics, the National Safety Council, and other sources.)

1,900: strep throat
3,000: food poisoning
5,000: choking
6,946: accidental hanging
7,450: pedestrians hit by a car
7,740: obesity
12,316: pregnancy-related
20,108: inflammation resulting from food or liquids getting into the lungs
35,000: antibiotic-resistant bacteria
35,823: alcohol-induced deaths
36,336: falls
40,922: blood poisoning resulting from bacteria
47,173: suicide
55,672: flu and pneumonia
64,795: accidental poisoning
83,564: diabetes
121,404: Alzheimer’s
160,201: chronic lower respiratory disease
169,936: all accidental deaths
250,000: medical errors
599,108: cancer
647,457: heart disease

Lot's of debate to be had about definitions, what constitutes avoidable, etc. But each of these categories represent some context of probability of return to health, cost of treatment, avoidability, etc. Whatever we choose to do to address Covid-19 necessarily is made within this larger context.