Extraordinary claims require … evidence

19 minute read


Thank you to University of Michigan PhD Candidate Danny Geiszler for providing critical feedback for this blog. Views are my own.

In my previous pandemic-related posts I’ve repeatedly encouraged readers to not believe everything you read on the internet, including my own editorializations and explanations. I do my best to highlight resources of interest and give explanations of scientific concepts (like R0 for viral transmission) or quirks of the scientific enterprise (pre-prints on biorxiv versus peer-reviewed journals). Because my training lies in genetics and statistics, I’ve tried to avoid topics that are incredibly outside my wheelhouse, and I do the requisite reading when explaining topics I haven’t been formally taught. I care most about pointing people to resources where they can see the data for themselves, but I know it may be challenging for those without a background in these fields to interpret the scientific data … it’s still difficult for me after 7 years of postbaccalaureate scientific training! So my other goal is to give non-experts tools to gauge the veracity of sources: is this peer-reviewed?, what is the scientific consensus on these findings?, are methods clearly explained?, is this an established source?, who is funding this?

Another science communicator, Dr. Samantha Yammine, provides advice for vetting sources at the end of a video explaining some inaccuracies in the Bakersfield doctors’ press conference: “At the end of the day don’t believe something someone says just because they have fancy letters at the end of their name or they’re in scrubs. If they’re pushing credentials on you and being overly assertive, in my book, that’s a red flag. Next time you see someone like that, especially if it’s someone, I don’t care who, going against what so many public health experts are saying, please reconsider spending your social currency on social media sharing it. And if you end up deciding you do want so share it, before you do that, try looking up if the opposite thing is true. Don’t look for what individuals are saying, try to get a sense of what consensus is, what majority of people are saying. While it is sometimes true that an individual is correct against all of the majority, it’s pretty rare, and if their evidence is spotty they’re probably not the exception.” If you’d like to learn more, North Carolina State University Libraries have a video about evaluating source credibility. I found two imperfect tools to gauge website credibility and bias–NewsGuard and Media Bias/Fact Check. It turns out, they are used by the University of Michigan School of Information Center for Social Media Responsibility to calculate an ‘Iffy Quotient’, or the percent or URLs with iffy credibility, shared on Facebook and Twitter at a given time.

This brings me to the credibility of a video which was circulating social media last week called “Plandemic.” It caused a stir with claims about aspects of the current pandemic and alleged controversies regarding HIV/AIDS and Ebola. I will not be linking to the video so as not to drive traffic to what many scientists and doctors believe to be a dangerous piece of misinformation from an uncredible source, although a full transcript is on Medium. Lots of people have done the necessary work to provide evidence refuting the video’s claims. To be clear, I do not think scientists and medical professionals are infallible, I acknowledge that unethical behavior occurs, and I am not so naive as to believe those in power are immune to ulterior motives. But extraordinary claims require … evidence. Thus far, there is very little evidence supporting the video’s conspiratorial claims. Below is a synopsis of places to find information about “Plandemic” and I summarize some lessons learned for my scientist colleagues.

“Plandemic” fact checking resources

Several science journalists and others have attempted to verify the claims with evidence, publications, etc:

A few Facebook posts that have gained traction that attempt to provide alternative perspectives:

There was also a fact check by Dr. Mike with a YouTube video if you’d prefer video format.

Lessons for the scientific community

For scientists reading, I posit this is a unique time for us to take a more direct approach at
1) educating the public about our tax-payer funded research (COVID-19 related or otherwise) 2) engendering public trust by explaining who we are and our motivations and
3) empathetically correcting misinformation with scientific evidence when we are able.
Tara Haelle covered a lot of this in her piece for Forbes. I’ve reflected on some lessons for the scientific community based on the comments from my own Facebook post that garnered quite a lot of debate.

Some believers of “Plandemic” are those who engage in conspiratorial thinking. There is a rich body of peer-reviewed research on conspiracy theories. With a quick search I found this, and I’ve seen this recommended to better understand where some people are coming from. Showing empathy, affirming critical thinking, and avoiding ridicule are important here. More specific to the time at hand, this peer-reviewed work examined why people believe COVID-19 conspiracy theories. They found “beliefs in two popular variants of COVID-19 conspiracy theory are the joint product of the psychological predispositions 1) to reject information coming from experts and other authority figures and 2) to view major events as the product of conspiracies, as well as partisan and ideological motivations.”

Some people have internalized misinformation without recognizing it as such, and the “Plandemic” video further confirms what they’ve heard through confirmation bias. In 2018, a study showed falsehoods on Twitter spread faster, deeper, and broader than the truth. A recent article in The Atlantic by misinformation expert Renee DiResta, highlighted the need to combat misinformation: “If institutions and authority figures don’t adapt to the content and conversation dynamics of the day, other things will fill the void. The time for institutions and authorities to begin communicating transparently is before wild speculation goes viral”. Another piece in The Atlantic by Yale lecturer, Liz Neely, walks us through ways to help others be better informed about the pandemic. Start with people who know and trust you, pick your battles, affirm shared values, start the conversation broad without jumping to correct the misinformation, and be honest and transparent.

Some believers of “Plandemic” are not sure who to trust and may not have the prior knowledge to come to their own conclusions. To assist these individuals, here is a list of topics I believe we should try to clarify to non-experts when given the opportunity. Topics related to COVID-19 misinformation specifically are italicized. A lot of them deal with the systems and checks and balances that we know exist in settings across academic and government biomedical research, but are quite opaque to the rest of the world: the peer-review process, pre-prints, retractions, grant funding, study sections, academic hierarchies (phd candidate, graduate student research assistant, postdoc, assistant professor, tenure), salary, conflict of interest disclosures, technology transfer, personal motivations for pursuing science, salary, universities versus private research institutes versus governmental research organizations, the scientific method, the iterative process of science, necessary vs sufficient, correlation vs causation, institutional review boards, how ethics violations are handled, the organizational structure of NIH and NSF, replication studies, patents for biomedical research especially vaccines, the Bayh-Dole Act, clinical trials, pharmaceutical pricing, the role of pharmaceutical companies in research, the reason for pandemic preparedness exercises like this, the uncertainty around models including 95% confidence intervals

I observed common beliefs that provide some additional perspective to consider as we communicate moving forward:

  • Any association with mainstream media (although what is considered mainstream is heterogeneous amongst individuals) automatically discredits information that lies therein, even when it was formerly a trusted source. When presenting evidence, we may have to select new sources that a given individual trusts. For example,
    A: I’ll believe Plandemic when I see it on 60 Minutes
    B: 60 Minutes is produced by CNN and CNN lies all the time (Fact check: 60 Minutes is produced by CBS News)
    A: Really! I didn’t realize :(

  • “I believed this until I realized you worked for the government.” That the government is not trusted will come as no surprise, but you can read about 2019 Pew Research Center results about trust and distrust in American government.

  • “I don’t trust scientists.” Pew Research shows trust in scientists in on the upswing in America, but the non-random sample of the Facebook post comments represents many that don’t trust scientists. Perhaps proactively communicating our science and engendering public trust would bolster trust in science in times of crisis. Some also suggest that the general public needs to have their expectations tempered for what the experts can and can’t know. “We need to value scientists and listen to experts, but part of listening means understanding that right now, what they’re saying is that they do not have all the answers.” Maybe that starts with scientists making our limits of expertise more clear by normalizing the phrase “I don’t know.”

  • “The truth is somewhere in the middle.” I would posit that when 99 experts say position A and one expert says position B, the onus is on the one expert to provide unassailable evidence to move the needle anywhere towards middle ground. However, I saw this comment frequently, with many claiming neutrality with regard to claims in ‘Plandemic’. The existence of a middle ground is impossible when one position is a verifiable piece of misinformation. Although simplistic, if 99% of experts say the sky is blue and 1% say the sky is red, it doesn’t mean the sky is purple. Being aware of this common attitude could be informative in future conversations.

Q & A

It was obvious many people had genuine questions. Here are the few I had the time to tackle in case the answers are helpful for you in your discussions with others. If one had unlimited time, I’m sure they could be better researched or supported. For most facts I only provide one citation, but I believe finding credible sources that converge on similar findings is key.

Q: Why are we shutting down for a disease with 99.6% survival rate?
A: In terms of the survival rate for COVID-19, epidemiologists often talk about the inverse of this—-the infection fatality rate (IFR) which is the number of deaths of all infected cases, not just the ones we have found by testing or hospitalization. You can read more about IFR and case fatality ratio (CFR) as they relate to COVID-19 here. In summary, IFR is estimated for a given population and can increase due to overwhelmed hospitals, poor underlying health, older populations, etc. These rates can also decrease as medical professionals become more skilled at treating the disease. I say all this to indicate there is uncertainty involved in estimating what proportion of people die from COVID-19. Right now we aren’t testing enough people to know our infection fatality rate (IFR), because infections are happening that aren’t recorded as cases. As we perform seroprevalence surveys to test for antibodies, we can understand how many have been infected and therefore estimate IFR. Early work does point to a fairly low IFR in some populations but there are limitations to these studies and we need more data! But the early IFR estimates for COVID-19 are greater than the 0.004% IFR for the flu for example (here is a Twitter thread with a publication). If the IFR is 0.4% and we multiply this by the millions in the U.S. who would be infected (let’s say 50%, which is a conservative estimate of herd immunity, of our 328 million population) we would have 656,000 deaths over some period of time, maybe a year. The larger concern is that COVID-19 can spread exponentially in a population with no vaccine and no immunity from previous infections. This can overwhelm health systems resulting in more deaths and a larger IFR. Many of the states’ stay at home orders reduced transmission (as seen by decreased hospitalizations in NY and MI, for example) to provide time to prepare hospitals and increase capacity for contact tracing which will be very important!

Q: Won’t washing my hands so much remove the good bacteria on my hands and therefore weaken my immune system?
A: It sounds like you’re interested in the microbiome, I am too! Our skin microbiome has transient flora which we get from touching objects and resident flora which is what would be important for our immune systems. Soap, warm water, and friction can physically break down the virus, and you don’t need to use anti-bacterial soap to remove the SARS-CoV-2 virus from our hands. The normal bacteria that colonize our skin, what we call the skin microbiome, will remain intact through frequent hand washing because they’re in the superficial layers of our skin and can also regrow if some of it is removed from hand washing. You can read more here.

Q: What about carbon dioxide poisoning from wearing masks?
A: Here is a fact check by Reuters. The amount of CO2 the average person would breathe in from a cloth mask for time spent in public is not dangerous. I can see why many people find them hot and stuffy, I do too, but it’s important to me to protect others around me in the event I’m sick and don’t realize it. Even more restrictive face masks like N95 respirators (which the public is not being asked to wear) are used frequently in professions like construction and medicine. I searched PubMed for peer-reviewed research and did see several small studies had measured the elevated carbon dioxide in the “dead-space” of N95 filtering facepiece respirators (e.g here), but I saw no evidence that this is so dangerous as to not use them to prevent infection.

Q: How do you explain this research if chloroquine isn’t an effective treatment?
A: There are a few reasons why the findings in this paper can’t be directly applied to the current pandemic. It is about the original SARS virus and not the novel coronavirus, SARS-CoV-2, which causes COVID-19. This paper showed therapeutic benefit in primate cells (what we call in vitro), but didn’t demonstrate effect on a living human (in vivo) which can sometimes have different outcomes than studies in cell lines. The drug in this study is called chloroquine, it’s closely related to hydroxychloroquine. Both chloroquine and hydroxychloroquine are being used in randomized controlled trials, the gold standard of medicine to prove drug efficacy. A pre-print (not yet peer-reviewed) showed the drug did not reduce the risk of mechanic ventilation and there was increased death in those treated with hydroxyochlorquine. A peer reviewed observational study showed the drug was not associated with increased or decreases intubation or death. Study results are still coming out, but the early signs show this drug could do more harm than good by causing heart rhythm abnormalities. I know some early studies in small sample sizes and anecdotal evidence looked like hydroxycholoroquine was a promising treatment, but science is iterative and we update standard of care when we learn more.

Q: The Pirbright Institute has a US patent on the coronavirus and Vanderbilt owns a large number as well. See here.
A: Coronaviruses are a broad class of viruses with some causing the common cold and others like SARS-CoV-1 causing the 2003 SARS epidemic. The novel coronavirus, SARS-CoV-2, which causes COVID-19 is different from this patent. The link you provide regarding the Pirbright patent says: “DISCLAIMER: Coronavirus is a broad name for a family of viruses. This patent is NOT for the new COVID-19 virus and The Pirbright Institute does not currently work with human coronaviruses. If you share this patent online, be aware you are in fact sharing a separate patent for avian infectious bronchitis virus and porcine delta-coronavirus. This is not a patent for the new COVID-19 virus.” The world of patents in medicine is definitely a complicated one, so I understand how this might be confusing.

Q: Didn’t Dr. Fauci say they were using previous research and the existing coronavirus vaccines to jump start the process for finding a vaccine for the novel coronavirus? His comments suggested they were much closer to a new vaccine because of the previous work they had done. And this wasn’t originally a human coronavirus, was it?
A: Previous research on coronaviruses has been really helpful in the effort to quickly create a safe and effective vaccine to help curb the spread of COVID-19 across the world. That being said, I believe in vaccine safety and clinical trials are underway to ensure even the vaccines produced are safe. Because the previous SARS pandemic in 2003 and MERS outbreak in 2012 were both from types of coronaviruses, many scientists have been aware of the continued risk of coronaviruses jumping from animals to humans. This is called a spillover event, and this page has a bit more information. You can read a comparison to SARS-CoV which caused SARS and SARS-CoV-2 which causes COVID-19. Because of the inevitability of another coronavirus outbreak, scientists have been working on vaccines that can target these types of viruses. The NIAID has a nice video from January explaining their vaccine research (many organizations across the world are working on vaccines that have different mechanisms). This article also touches on the fact that because of their work with MERS, scientists know how to optimize vaccines that will target the spike protein seen on the coronavirus that causes MERS and the coronavirus that causes COVID-19. I hope this also answers your question regarding if this was originally a human coronavirus. The exact origins of the virus have not been confirmed, but scientists are confident that is has natural origins with a spillover event from an animal like previous coronaviruses. You can read more here, here, and here.

Q: A secondary interview with one of the Bakersfield doctors seems to clarify their original points, what do you think?
A: Here is the joint statement that was issued by two professional organizations for emergency medicine. It has an additional information section which describes how difficult it is to compare flu versus COVID-19. I disagree that there isn’t data supporting social distancing and isolation. Peer reviewed work in a well respected journal, JAMA, showed that non pharmaceutical interventions (NPI, the technical term for mitigation strategies like social distancing) reduced the Rt (average transmissions one infected person makes) substantially in China (Figure 4). A pre-print looking at 20 countries showed the reduction in number of new cases from a variety of mitigation techniques like quarantine. We even know social distancing worked in the 1918 pandemic. There are many popular science articles about this, but I found a peer-reviewed article from before COVID-19.
He states there are 14 patients hospitalized in his county for flu, 3 on ventilators. I would need to know how many were hospitalized with COVID-19 on the same day to put those numbers in context with the point he is trying to make. I tried to make a similar comparison from their public health website. The county had 9 deaths for flu in 2018-2019 season and has already had 15 resident (10 if you count non-resident) deaths in the first few months of the pandemic. We know the flu has demonstrated seasonality, but we don’t yet know if COVID-19 will have seasonality meaning it could continue to move through an immunologically naive population (without a vaccine like we have for flu) all summer making for a much longer “season”. Here is a pre-print discussing projections for the summer. But because the flu and COVID-19 are still so different (length of treatment longer with COVID-19, difference in way of counting deaths, constellation of complications presenting with COVID-19) I hesitate to put much weight in these comparisons.
The ultimate goal is herd immunity, but no one is advocating for shelter in place until then. As he mentions, there is collateral damage of shutting down the economy—-the public health ramifications from economic downturns are real and weigh heavily in all of our minds. From my reading, I believe most epidemiologists would say the strategy for a country like the U.S. which had such widespread community transmission is to reduce Rt with suppression strategies (e.g. shelter in place) before carefully re-opening and beginning test, trace, isolate. This prevents health care systems from being overwhelmed, provides time for PPE manufacturing and therapeutic development, and builds confidence in consumers for participating in the economy, etc. Then we reach herd immunity by combination of the virus moving more slowly through the population and a vaccination program when one is available. If we reach herd immunity by letting the virus run through the population, by assuming Rt of 3 and a fairly low infection fatality rate (IFR) of 0.2%, that would mean ~350,000 deaths in the U.S.(Figure 2C). Scientists are in the process of figuring out a more accurate IFR through seroprevalence studies, but I’ve seen early estimates around this value. I acknowledge the model’s assumptions are not perfect and are being revised and replicated by different groups. You can see in that figure that Sweden has a much lower predicted mortality than the US. I think this opinion piece describes my position on Sweden pretty well—-there’s a lot we still don’t know!

Q: Didn’t Dr. Judy Mikovits worked for Dr. Fauci because he is at the NIH and so was she? A: I cannot find Dr. Mikovits’ CV online, but according to her Wikipedia page and the affiliations I see on her published papers, she never worked at National Institute of Allergy and Infectious Disease (NIAID) where Dr. Tony Fauci is the director. As I stated in my original post, NIAID is one of 27 institutes and centers of the National Institutes of Health (NIH). It appears the bulk of Dr. Mikovits’ work was at a different institute, National Cancer Institute (NCI), and a private research institute, Whittemore Peterson Institute. Therefore she would have never been in a direct reporting line of Dr. Fauci. I also provided as a caveat, it’s possible that she received extramural grants from NIAID at some point in her career. Regardless, in the world of science you only “work for” someone if they are the principal investigator (PI) of your laboratory or you’re a PI and they are a departmental/branch chair. If you work for someone you’d likely be on papers with them as co-authors, and a search of PubMed shows Dr. Fauci and Dr. Mikovits are not co-authors on any publications that I can find.