Today I received a post written by Aaron Ginn titled “Evidence Over Hysteria – COVID-19” from friends who are using it to believe we are overreacting and they don’t believe the threat is nothing worse than the seasonal flu. I’ve also seen it posted a few times on Facebook from others.
(Note: I wrote this quickly so please forgive its disjointed nature and lack of well-constructed grammar.)
To sum up the author’s article, he is attempting to use data to let everyone know the MSM (and scientists, epidemiologists, doctors and medical professionals) are wrong. We are all overreacting. We shouldn’t be closing schools and implementing shelter-at-home policies.
He wrote a great narrative and it sounds very convincing to those who are predisposed to believing his position as it casts aspersions and does a great job of planting doubts in peoples’ minds. Not only is this piece irresponsible at this given moment, it’s also dangerous.
Please keep in mind: he’s not a doctor, scientist or epidemiologist. He’s a young, self-described “growth hacker” who has grown up working in conservative politics. He has no scientific credentials or credibility. On the same note, I am none of the above either and I also have no scientific credentials or credibility.
So consider this a rebuttal from one person with no credibility to another person with no credibility. If people were so quick to latch on to his article to support their view, maybe others will latch on to my rebuttal also in the same manner.
In fairness, he did put this at the bottom of his dissertation, “You may ask yourself. Who is this guy? Who is this author? I’m a nobody. That is also the point. The average American feels utterly powerless right now.” I can relate to him on his last sentence as I feel most Americans can agree, especially right now.
If you choose to read, please form your own opinion and, most importantly, don’t listen to me or him. Listen to the scientists and experts. We are both nobodies.
In his article, one of his main points is that we’re looking at the data wrong. He writes, “Still, there is a massive blind spot with this type of graph. None of these charts are weighted on a per-capita basis. It treats every country as a single entity, as we will see this fails to tell us what is going on in several aspects.”
His next sub-headline states, “On a per-capita basis, we shouldn’t be panicking.” He supports that statement with this graphic:
He purports that we’re looking at nominal (total) cases when we should be looking at the number of cases per million people. It sounds convincing to people, “Oh, he’s right. We only have 59 cases per MM people where Italy has 778 and Spain has 461. This is great news! Nothing to worry about!”
This is the first of a few misdirectional arguments. For time’s sake, I won’t cover the obvious first issue relative to the US being approximately two weeks behind the outbreak in Italy and Spain which is why our numbers are lower per population. The second issue is testing. It’s impossible to have positively identified cases when we have no way to test for it and thousands of people with symptoms are unable to even qualify for testing.
The third aspect of his argument is important to cover. He is right in that the number of cases per capita is a very informative statistic relative to how much the virus has penetrated a country. But he then attempts to link that statement to “we’re doing better” than Italy, Germany, France, Spain, etc. It’s a false equivalence.
The right way to look at the data is not by looking at total cases divided into the total population but rather to view the data relative to population density.
Simply put: the closer people are together, the greater chance the virus will spread faster. In a dense, closely populated area the effective R0 (pronounced as R naught) will be higher and the virus will spread faster than in a rural, sparsely populated area. Common sense, right?
To help explain, below is a more accurate way to think about it relative to population data. I’ll use Italy as our proxy and then include a data table for similar countries.
The size of Italy (sq km) is 60,462.
The size of US (sq km) is 7,895,310 (I removed Alaska from my US numbers to be fair and more accurate; with Alaska our number would be 9,372,610).
Population of Italy: 60MM
Population of US: 331MM
Population Density (Italy): 200 ppl/sq km.
Population Density (USA): 42 ppl/sq km.
Summary: Italy is 4.7x MORE population dense than America.
If the US was as densely populated as Italy, we would have 1.6 billion people in our country which would make us, by far, the largest country in the world. Carrying the analogy forward, assuming a virus spreads in relative proportion to the social proximity (population density), we would expect the virus to have both a higher R0 and, as a percentage of the population, for it to penetrate almost 500% faster (which is exactly what it’s doing as an overall percentage of the population in more densely packed countries).
The right way to think about the spread is to view the US, not as one country, but rather as 50 separate little countries. In the most densely packed cities (and states) the virus is going to spread at a much faster pace than in rural, sparsely populated areas. Hence why NYC, our most dense city, has become the epicenter with Chicago, Boston, Newark, San Francisco and Seattle close behind.
The premise of his argument that we are much lower in cases per capita is built off incorrect assumptions including time, testing and population density.
Even though his argument is faulty and his position of not doing anything is dangerous, there is a silver lining for the US. Because of the sparseness of our population, it serves as a natural and geographical inhibitor to an outbreak at the scale we’re seeing in Italy, Spain, etc. Our densely populated cities will, because of urban density, get hit the hardest.
The other two items working in our favor are: time and weather. Because we’re a few weeks behind, we have the ability to learn and implement smarter policies to get in front of the contagion. The second, although the data isn’t statistically relevant yet, is that it appears warmer, more humid weather may be an inhibitor to the virus spreading.
Between our geographic dispersion and time on our side as we move into Spring and warmer weather appears, we most likely have two natural defenses working in our favor.
The measures many governors and mayors are taking to shut down and isolate people to get it under control are critical to get the R0 to R1 or less. Restriction of movement is the only thing that can slow it down and that is the #1 goal. There is no other option.
The more we can slow it down, the more “time we buy” for many things including:
- Hospital and healthcare facilities preparedness
- Protective equipment: masks, gloves, body suits
- Medical equipment: ventilators, breathing machines
- Therapeutic Drugs
- Vaccines
I’m sure many are already tired of hearing “flatten the curve”, but it’s critical. Listen to the experts and what they’re telling us. Don’t listen to me or a political growth hacker.
Thanks. Stay safe. Stay informed. Stay inside.
I came back to his post this morning to include a few additional byte-sized comments and thoughts:
After his first point above, his second point is, “COVID-19 is spreading, but probably not accelerating” and, “Daily growth rates declined over time across all countries regardless of particular policy solutions, such as shutting the borders or social distancing.”
I’m not sure what he’s talking about. Every single country has implemented some form of policy solutions. There is no country that’s been minimally effected that hasn’t implemented policy solutions. He tries to get the reader to believe that, if we did nothing, growth rates will naturally decline and that shutting borders and social distancing have no effect on the growth or decline of the virus.
Although the majority of his article is filled with misdirection through the use of data, his next statement is a doozy:
“Cases globally are increasing (it is a virus after all!), but beware of believing metrics designed to intentionally scare like “cases doubling”. These are typically small numbers over small numbers and sliced on a per-country basis. Globally, COVID-19’s growth rate is rather steady.
Viruses though don’t grow infinitely forever and forever. As with most things in nature, viruses follow a common pattern — a bell curve.”
This is a ridiculous statement. The obvious response is, “no kidding”. By definition, because populations are finite numbers, the number at some point will start to decrease and will represent some form of a bell curve. It’s inevitable. The only question is the numbers inside the bell curve. Will there be 1MM, 10MM or 100MM people inside the bell curve? That is the number we’re going to work hard to keep under control.
At some point, by definition, the number will start to decrease. Forget about doubling, look at the raw numbers by day (to be fair, this is more a case of finally testing people which we were late on versus a representation of actual new infections. The reporting data is at least a week or two behind the reality. Once we start catching up with tests, combined with our social distancing and quarantine policies, the daily numbers will start to decrease.)
Then we move to this headline, “A low probability of catching COVID-19”.
His statement, “According to their report if you come in contact with someone who tests positive for COVID-19 you have a 1–5% chance of catching it as well.”
Once again, using data for misdirection purposes. The above statement is true. But let’s go back to the population density discussion.
I’ll use an example: let’s assume I come in close contact with 200 people in the course of a week (I have 700 people at my company and spend time in a lot of airports, restaurants and bars; this number may be an understatement). If 5% of them are infected, that’s 10 people I came in contact with. Multiple the 1-5% by 10 and I’ve got anywhere from a 10% chance to 50% chance of contracting the virus. And that’s in one week. Imagine how high my odds would be over the course of 2-3 months.
One of his next headlines is, “1% of cases will be severe.“ He writes, “Looking at the whole funnel from top to bottom, ~1% of everyone who is tested for COVID-19 with the US will have a severe case that will require a hospital visit or long-term admission.
First, it’s way too early to even postulate on the above statement as we’ve only tested around 20,000 people. Second, worldwide data shows severity of people tested is at least double, and most likely triple, the 1%. And then he follows up with his statement and shows the chart below. Maybe he assumes his readers can’t understand a basic pyramid chart?
Looking at global data, here’s what we can postulate although I feel it’s way too early to even begin trying to use the limited and inaccurate data to draw any conclusions: of the people tested, approximately 20% have a positive identification rate. Of the 20%, approximately 15%-20% of the cases are considered “severe”. That would put the severity of tested people closer to 3% to 4% versus 1%.
The reality is we don’t know enough and we don’t have accurate enough data to make any broad or sweeping generalizations or predictions. We aren’t testing anywhere near enough, the reporting of the tests are incomplete and inconsistent and it’s in the early stages.
Giving him the benefit of the doubt, I’ll assume his intentions are well-intended even though they are highly inaccurate, misguided, misleading and dangerous given what we don’t know. Personally, I’m of the opinion we move forward with an abundance of caution and, if we’re going to err, let’s err on the side of being overly conservative in our projections and implement programs to thwart the spread of COVID-19. Social distancing, at this moment, is the only proven way to limit the virus from overwhelming our healthcare facilities. To purport or recommend anything different is grossly irresponsible.
I do want to offer support for one of his last sentences, “The health of the State will be even stronger with more Americans dependent on welfare, another trillion stimulus filled with pork for powerful friends, and a bailout for companies that charged us $200 change fees for nearly a decade. Washington DC will be fine. New York will still have all of the money in the world. Our communities will be left with nothing but a shadow of the longest bull market in the history of our country.”
This is the sin of all American crises where the average, and in this case, the most economically vulnerable of our citizens, will get raked over the coals. The corporations and political parties always seem to come out ahead while the common person gets screwed.
My hope is this crisis opens up the door for a reset. The financial disparity between the wealthiest in our society and the corporations has achieved a level of unsustainability. This crisis will expose and lay naked the societal ills and economic depravity political leaders have foisted upon the majority of working Americans for the past 50 years.
Things have to change. Let’s get through this crisis first and then we need to make some serious changes for our future.
