Dan Erickson, screenshot
Dan Erickson owns seven urgent care facilities in California. He recently produced a viral YouTube video in which he complained that business volume in his seven facilities is down by half, due to mandatory social distancing in his area. In his video, he attempted to persuade viewers that it is time for them to stop wearing masks and to demand an end to social distancing.
We need to notice something about character here. In his video, Erickson deceitfully claimed that ER doctors, along with the head of the California Department of Public Health, and the head of the local Kern County Health Department agree with his position, stating, "[W]e're all in agreement..., they're in agreement with me."
Since his video was posted, the California Department of Public Health has reiterated the need for social distancing, and the Kern County Public Health Department responded to a reporter's question on Thursday--
Q: I was wondering, the two doctors yesterday from Accelerated Urgent Care said they spoke with Mac Constantine [director of the Kern County Public Health Department], and he agreed with them about maybe a need to reopen soon. Is that something you can speak on? They said that he said he's following the state's orders, despite his agreements with them.
The American College of Emergency Physicians (ACEP) and the American Academy of Emergency Medicine (AAEM) responded more sharply in an usual joint statement--A: So, I can confirm that the doctor at Accelerated has spoken with our director. Our director has not concurred with the statements that were made yesterday about the need to reopen at this time. What I would like to reiterate is we take very seriously the governor's guidance, the guidance from the California Department of Public Health and the Centers for Disease Control, and we want to reiterate: Now is the time to remain very vigilant and stay at home and practice social distancing. We know that if we practice social distancing this is the one thing we can do to flatten the curve in our community and not burden our healthcare system, so I strongly state that that is our position. [voice emphasis in original]
[We] jointly and emphatically condemn the recent opinions released by Dr. Daniel Erickson and Dr. Artin Massihi. These reckless and untested musings do not speak for medical societies and are inconsistent with current science and epidemiology regarding COVID-19. As owners of local urgent care clinics, it appears these two individuals are releasing biased, non-peer reviewed data to advance their personal financial interests without regard for the public’s health.
COVID-19 misinformation is widespread and dangerous. Members of ACEP and AAEM are first-hand witnesses to the human toll that COVID-19 is taking on our communities. ACEP and AAEM strongly advise against using any statements of Drs. Erickson and Massihi as a basis for policy and decision making.
Dan Erickson had the medical degree, the scrubs, the stack of papers, and plenty of numbers to rattle off. He also claimed that everyone of importance agrees with him. That made him seem credible, but his reasoning was extremely sloppy.
As I demonstrated yesterday, he used the Fallacy of Free Extrapolation to dupe his viewers. But that is not the end of his specious claims and sloppy errors. I would like to talk about the concept of statistical significance.
As I demonstrated yesterday, he used the Fallacy of Free Extrapolation to dupe his viewers. But that is not the end of his specious claims and sloppy errors. I would like to talk about the concept of statistical significance.
I know, right?
But the basic concept is very simple. When two groups are fairly small or the difference between the two groups is fairly small, we sometimes can't tell if the result is due to chance or due to some other factor. The difference may not be statistically significant. To make a qualitative statement about statistical significance, we don't need to know what caused something. We only need to know whether the difference between two groups is so small that it could be due to chance or so large that it must be due to something else.
Chance has the biggest chance of producing a measurable difference in little groups with small differences. Flip a quarter ten times, and you may get heads six or seven times. With very big groups and very big differences, however, it quickly becomes impossible to assign a difference to chance. Flip a quarter five hundred times, and the result won't be 60% heads.
A bigger difference between bigger groups will always be more statistically significant than a tiny difference between tiny groups. Let's consider two neighboring streets, Pine Street and Oak Street. With a similar demographic distribution on both streets, there are eleven cases of COVID-19 on Pine Street and nine cases on Oak Street. A statistician would work out the math for us and tell us that the difference is not statistically significant because this is a very small difference between two very small groups.
But let's consider COVID-19 in New York and Wyoming. New York has 1153 deaths/million. Wyoming has 12 deaths/million. Is this difference due to chance? Of course not. Various factors are present in New York that are not present in Wyoming, and we can identify several of them. The big difference between these two big groups is statistically significant.
When Dan Erickson talked about statistical significance, he did not ask us to look at a small difference between small groups. He directed our attention to a massive difference between two massive groups, saying,
But let's consider COVID-19 in New York and Wyoming. New York has 1153 deaths/million. Wyoming has 12 deaths/million. Is this difference due to chance? Of course not. Various factors are present in New York that are not present in Wyoming, and we can identify several of them. The big difference between these two big groups is statistically significant.
When Dan Erickson talked about statistical significance, he did not ask us to look at a small difference between small groups. He directed our attention to a massive difference between two massive groups, saying,
When you bring up a system of lockdown, you automatically have to compare it to a system of no lockdown. Sweden and Norway.... Norway has lockdown. Sweden does not have lockdown. What happened in those two countries? Are they vastly different? Did Sweden have a massive outbreak of cases? Did Norway have nothing?
Let's look at the numbers. Sweden. Sweden has 15,322 cases of COVID. They did 74,600 tests, which is 21%...of all those tested came out positive for COVID. What's the population of Sweden? About 10.4 million. So, if we extrapolate out the data, about 2 million cases of COVID in Sweden. They did a little bit of social distancing. They would wear masks and separate. They went to schools, stores were open. They were almost about their normal daily life with a little bit of social distancing. They had how many deaths? 1,765....
"Norway. Its next-door neighbor.... These are two Scandinavian nations. We can compare them, as they are similar. Let's look at the data. Norway. 7,191 cases of COVID. Total COVID tests 145,279. So they came up with 4.9% of all COVID tests were positive in Norway. Population of Norway, 5.4 million. So, if we extrapolate the data as we have been doing, which is the best we can do at this point, they have about 1.3 million cases. Now, their deaths as a total number were 182, fairly small, but statistically insignificant from 1700, you realize.
"Seventeen hundred. One hundred. These are statistically insignificant."Millions of cases. Small amount of death. Seventeen hundred. One hundred. These are statistically insignificant.
This statement is worse than nonsense. Dan Erickson has the education to know that he is telling a fib, and he is apparently expecting that reporters and viewers wouldn't catch him in it, any more than he expected to be caught in his statements about widespread physician and public health agreement with his opinions.
But we'll catch him. Before we do, let's give Dan Erickson a gift for his argument. The ratio of deaths was not 1700 to 100 between Sweden and Norway. It was 1765 to 182. As an absolute number, deaths in Sweden are almost ten times greater than deaths in Norway, not seventeen times greater.
And, let's give Erickson a second gift. The population of Sweden is nearly twice as big as the population of Norway. On April 21, the day that Erickson obtained his numbers, the death rate per million was five times greater in Sweden than in Norway. Death rate is more meaningful than an absolute count when comparing whole populations, so we'll use that instead. So it's five times greater. Not ten times. Not seventeen times.
Erickson insisted, "Seventeen hundred. One hundred. These are statistically insignificant."
But we'll ask a more reasonable question that is better for his argument. On April 21, Sweden had 173 deaths/million. Norway had 34/deaths per million. The death rate in Sweden was five times greater than the death rate in Norway. Is that statistically significant?
Are Norway and Sweden tiny groups or big groups? They are big groups.
Is 500% a tiny difference or a big difference? It's a big difference.
Could this result be due to pure chance? Not in this universe.
"Seventeen hundred. One hundred. These are statistically insignificant."
That's not an ignorant statement. It's deceitful manipulation from a deceiver.
Don't reserve all your skepticism for scientific consensus. Save some skepticism for the fringe, which also needs a good measure of accountability.
Stay safe and keep others safe!
Valerie
Comments
Post a Comment