Is the World Also Fighting a Propagandemic?

A review of two California doctors’ deceptive presentation on COVID-19

Sorting quality information from propaganda (a.k.a. bullsh*t with an agenda) is crucial to maintaining a free society. Right now it’s also key to making policy decisions to promote public good. That involves weighing costs and benefits to public and economic health in planning next steps in the pandemic response. Unfortunately, a recent presentation by Bakersfield, CA urgent care doctors Dan Erickson and Artin Massihi fails in this regard. People sharing the video of it on social media (update: YouTube and Facebook had taken it down as of 4/28/20) seem to perceive it as a reasoned case made in good faith. I take issue with both their statistical methods and projection of smug certainty. Their manipulation of “facts” and “science” to promoting personal and public health policies could cause great harm.

Here are some specific reasons for concern and my thoughts on why, in chronological order:

(3:30) “We’ve never seen where we quarantine the healthy.”

This statement is misleading. Healthy Americans aren’t quarantined unless by choice. They are using social distancing and mostly staying home per governors’ orders or instructions. There are plenty of historical examples of orders to “shelter in place” to protect people from threats outside of their homes, schools, or workplaces.

(5:00) “As you know the initial models were woefully inaccurate. They predicted millions of cases of death. Not of prevalence or incidence, but death. That is not materializing.”

It would be helpful to specify at least one model. Also crucial is knowing over what timespan, under what conditions (e.g., no social distancing, partial, or aggressive) and for what population the deaths were predicted.

(5:10) “What is materializing in the State of California is 12% positives.” “If we just take a basic calculation and extrapolate that out, that equates to about 4.7 million cases” and “a 0.03 chance of dying” from COVID-19 in California.

They “extrapolate out” based on the tests that have been done. For this to be legitimate, they must show that the tested population is representative of the larger population. Otherwise, there is sampling bias. If tests were given mostly to people with specific symptoms rather than a random sample, we could expect the number of positives to be higher than for the general population.

Their numbers are FAR lower than those in a study by epidemiologists examining cases outside of China that calculated a 1.4% case fatality rate for people younger than 60 yo and 4.5% for those over 60 (for cases outside of China). One reason why could be the doctors overestimating the prevalence through faulty extrapolation.

(6:15) “96% of people in California who get COVID recover with almost no significant sequelae or almost no significant continuing medical problems.”

It’s too soon to know about any long-term health implications of COVID-19, since it’s only been around for a few months. Instead he could say we just don’t know yet. There does appear to be potential for reinfection, and there could be damage to a variety of organs. Dr. Noah Nesin (Penobscot Community Healthcare) noted in direct communication that “a 4% rate of mortality and severe complications” is considered “catastrophic.”

(7:20) “39% of New Yorkers tested positive for COVID-19.” This is based on 256,272 positives compared to 649,325 tests. “They likely have 7.8 million cases in New York.”

In addition to the sampling bias issue discussed for California (above), if the testing was mostly done in New York City and surrounding areas, it seems extra inappropriate to extrapolate to the entire (relatively suburban or rural) state. Also of note is that the doctors’ estimate is three times that indicated by an antibody test conducted last week using 3,000 randomly selected subjects in different parts of the state. Researchers found 21% of NYC residents had antibodies versus 3.7% in upstate NY.

(8:20) “Some of [the models] were based on social distancing and still predicted hundreds of thousands of deaths, which has been inaccurate.”

He again provides no model names or sources of them. Assuming he’s referring to hundreds of thousands of deaths for the U.S., based on what level of social distancing and other factors? This type of vague generalizing is similar to criticism of “the media,” as though it is monolithic. We need more detail to promote accountability.

(9:00) NY has 19,410 deaths out of a population of 19 million people, “which is a 0.1% chance of dying from COVID in the state of New York.”

Assuming the death count is close to accurate, there WAS a 0.1% chance of dying of COVID in New York IN THE FIRST FOUR MONTHS of 2020. It is important to note the time frame used and the fact that the probability will increase over time as more deaths occur. This is also not a normal way to assess virus severity. Case fatality is the % of people who get the disease who die. Preliminary data from antibody testing yielded an estimated 0.5% case fatality rate for the state. It’s also worth noting that COVID-19, heart disease, and cancer are the three leading causes of death in the country, with the order varying day to day.

It’s further confusing for the doctors to present two types of death risk indicators. The number of those infected who die is the case fatality rate. The deaths within an entire state population over a given time period is distinct from that. It also depends on both virulence and time: it was zero when the virus began and will continue growing as more die from it.

(10:30) “We always have between 37 and 60,000 deaths in the United States. Every single year” (from the flu). “It’s even as low as 20,000 some years.” “And we don’t necessarily report all of our flu tests.”

The CDC-reported average (mean) over the last 9 years is 37,500 annual flu deaths in the United States. These numbers are estimates that account for doctors not reporting all flu test results. He compares COVID-19 to the 2017–2018 flu (the worst year in the last 9) rather than an average flu season (e.g., from the last 10 years). This is also a comparison of 3 to 4 months with COVID-19 versus 12 months with influenza.

(11:45) He calculates the risk of dying from COVID-19 in Spain as 0.05, based on 22% of those tested testing positive. He estimates 10 million cases and divides by 21,282 deaths.

This is problematic based on extrapolation from urban to suburban or rural populations, as noted above.

(12:50) Comparing Sweden and Norway, he says “Norway had lockdown, Sweden did not have lockdown.”

Sweden has reported 217 deaths per million citizens and Norway 37 deaths per million citizens. He later clarifies that Sweden has engaged in certain distancing, though Norway’s response has apparently been much more aggressive. He doesn’t mention the timing of the lockdowns or other measures, which greatly influence their impact. For instance, early action in Taiwan led to there being 6 deaths (0.3 deaths per million residents) there.

(14:35) “So if we extrapolate the data as we have been doing, which is the best we can do at this point.”

We can in fact do better than extrapolating without accounting for population differences across different countries. We can explain how the figures are subject to change, and that at this point we can still only draw tentative conclusions.

(15:35) “Child molestation is increasing at a severe rate.” “These things last a lifetime.” “This is from me talking to ERs, talking to my doctors, and talking to people across the country and finding out what they’re seeing.” “Suicide is spiking.”

Personal observation by doctors is useful, but we need more research to support his claim that something is “increasing at a severe rate” or “spiking” within a given population. He provides nothing subject to peer review, nor does he call for it.

A complete accounting of secondary effects of COVID-19 also includes positive ones, such as from decreased air pollution and fewer car accidents.

It’s also worth noting that death from COVID-19 or other long-term health complications (e.g., tissue damage, including from strokes among 30 or 40-year-olds infected with COVID-19 also “last a lifetime.”

(20:30) “Sheltering in place decreases their immune system.”

He seems to be describing COVID-19 guidance as including not going outside. In some Italian cities people have been told to stay inside. That’s not part of guidance for Maine.

There is ongoing debate within the field microbiology about “hygiene theory” and how to best promote overall and microbiome health. Staying indoors COULD suppress immunity over time. And certain microbe avoidance measures COULD have adverse immune impacts if employed in the long term. The question is how significant are the short-term effects compared to those of uncontrolled spread of a pandemic? Do the effects vary with level of immune system development? These are crucial questions for researchers to better understand. He fails to call for that or mention controversy about them in the field, instead presenting them as based on scientific consensus.

(27:45) “Those are fomites and carriers of disease, so you take your family sheltering in place, and you think it’s safe.”

He compares risk of buying potentially contaminated plastic products (a.k.a. “fomites”) at the hardware store (and “mingling” there - definitely not part of any official guidance) to going to church or work. Then he states that “you think you’re protected” by staying home, and that this represents “lack of consistency” in our response.

This appears to be a straw man argument. Only microbe illiterates would assume their homes were COVID-free due to mostly staying home. Which is why guidance includes washing our hands frequently, not just after going out. The point is to mitigate risk, not eliminate it. And reducing exposure to droplets is the most effective way we know of right now. So buying a plastic shovel confers a lot less risk than being at church in a confined space with a large group of people for over an hour.

(29:40) We are saying that our response now, now that we know the facts, it’s time to get back to work. It’s time to test people.”

We have more data every day. We will never know all of “the facts.” The question is with what confidence can we make certain assertions about infectiousness and virulence? They are far more confident than most virologists and immunologists. They do raise some interesting questions, but by no means is this a time to drop any mics.

(30:10) “You don’t talk about one thing. You talk about co-morbidities… COVID was part of it. It is not the reason they died, folks.”

Assessing a single cause of death is no doubt difficult. Co-morbidities matter, both for flu and COVID-19 deaths. If the CDC is applying different standards to one versus the other, it would no doubt make comparisons more difficult. Are they? He doesn’t provide evidence of this.

Determining death rates for COVID-19 is a highly challenging task. He fails to properly explain this. Many people are likely dying at home without being tested, and this number increases during virus peaks in places in which healthcare resources have been overwhelmed. Other causes of death become tougher to track as well at such moments. Death rates in general have gone way up throughout the world during the pandemic.

One thing Erickson and Massihi don’t talk about: other indicators of viral severity. For instance, extended hospitalization is a significant cost. COVID-19 patients’ average hospital stays may be double that for flu patients.

(30:55) On being “pressured” to “add COVID” when writing up death reports: “Why is that… to maybe increase the numbers and make it look a little worse than it is? I think so.”

Are there other possible explanations for why hospitals would want to account for COVID-19-related deaths? I think so. One would be to be able to trace the progression and impacts of the virus. He’s clearly insinuating that COVID-19 death counts are inflated, at least in the U.S. He fails to mention factors contributing to under-counting at this point.

(31:05) Dr. Massihi: “In order to open the economy, you have to have widespread testing. That’s number one, no question about it.”

This makes sense for those advocating for ending lockdowns. It’s unfortunate that they don’t lead with this or mention it sooner in order to not confuse people about national priorities for making re-opening possible.

(33:50) “We develop herd immunity. We develop the ability to take this virus in and defeat it. And for the vast majority, 95% of those around the globe, this is true.”

We don’t actually know if this is true. We can hope so, but it’s too soon to know what level of immunity those who’ve been infected have afterwards, and for how long. A significant level of protection is key to herd immunity.

(29:00) “We’re not wearing masks… We want strong immune systems.”

(34:25) On wearing masks or gloves: “Do you want your immune system built or not built?”

He’s conflating immediate risk mitigation and long-term healthy lifestyle approaches. Masks and gloves are designed to be short-term solutions to control an outbreak. Doctors regularly prescribe a course of antibiotics for bacterial infections. This can save lives but is a short-term solution. And it would be reckless to generally prescribe long-term use of antibiotics, including since they are known to disrupt gut bacterial balance.

Conclusions:

Economies and public health have suffered greatly from COVID-19. Our next steps are crucial for improving outcomes. The most successful re-opening strategies will likely come from drawing conclusions using the best data available of potential health and economic costs and benefits. That precludes deceptive statistical manipulation. These doctors appear to underestimate the costs of continued high levels of infection without a vaccine. They also overstate their certainty. This may garner lots of conservative media attention, but it degrades the quality of debate and resulting policy.

Random questions about Dr. Erickson:

Is he a sociopath? He repeatedly says “millions of cases, small amount of death,” without any sign of compassion for those suffering (including in hospitals for long periods of time) or who have lost loved ones. Instead he smiles or smirks throughout the presentation and in response to questions. Is there a trace of sympathy or empathy in any part of the presentation? (11: 45) He flatly describes the race for the most cases as one “we don’t want to win.” (14:45) Comparing Norway and Sweden he points out while smiling that 182 deaths in Norway is “statistically insignificant” from 1700 in Sweden. What’s up with that, Doc?

Is he seeking some national political appointment? (9:40) He discusses U.S. testing and uses the word “impressive.” Agreed that 4 million tests is a large number. He makes no mention of the lag in testing compared to countries like South Korea or the need to have so much in part as a RESULT of failure to control infections early on (including through testing). He also mentions multiple times that he is from Norway, which appears to be the POTUS’s favorite European country. In addition, both doctors appeared on Laura Ingraham’s “The Ingraham Angle” on Fox News on April 27.

Did he get food poisoning after eating at Del Taco once? (28:35) It’s pretty messed up to name them in his example of food prep that could transfer the virus. Although ideally all food workers are practicing effective safety measures. Preparing food with gloves also in contact with money or other potential sources of contamination is dangerous at any time.

Does he have stock in Equinor ASA (formerly Statoil)? (15:05) He asks does the difference in loss of life between Norway (home to Equinor ASA) and Sweden necessitate “shut down, loss of jobs, destruction of the oil company, furloughing doctors?”

Thanks for reading and hopefully supporting my quest to promote quality information and debate.